feat(tools): added GitHub Actions workflow renderer and enhanced gh-* tool outputs

- Added gh-renderer.ts module with GitHub Actions workflow visualization for terminal output. - Enhanced gh_pr_view tool to fetch and display inline review comments alongside pull request reviews. - Improved gh_run_watch tool output with structured job state tracking and failed log formatting. - Fixed gh_run_watch to resolve explicit branch watches against GitHub API branch head instead of local HEAD. - Fixed gh_* tool outputs to spill full large results to artifacts instead of pre-truncating. - Standardized trailing newlines across 100+ prompt template files for POSIX compliance.
2026-04-01 17:29:47 +02:00
parent a0f2b0ad99
commit b7c801fab2
108 changed files with 1234 additions and 272 deletions
@@ -1,8 +1,10 @@
 # Changelog

 ## [Unreleased]
+
 ### Added

+- Added dedicated `gh-renderer.ts` module for rich terminal rendering of GitHub Actions workflow runs with live status snapshots and job details
 - Added `gh_pr_checkout` tool to check out GitHub pull requests into dedicated git worktrees with contributor push metadata
 - Added `gh_pr_push` tool to push checked-out pull request branches back to their source branches
 - Added `gh_repo_view` tool to read GitHub repository metadata using the local GitHub CLI
@@ -30,11 +32,20 @@
 - Added opt-in `gh_run_watch` to fast-fail GitHub Actions runs, stream job status snapshots, and return tailed logs for failed jobs
 - Added bundled `/green` command to generate the iterative “fix CI until green” prompt, with final tag instructions included only when `HEAD` already has a tag

+### Changed
+
+- Enhanced `gh_pr_view` tool to include inline review comments alongside pull request reviews for improved discussion context
+- Improved `gh_run_watch` tool output rendering with dedicated visual component for streaming run snapshots and job status updates
+
 ### Fixed

+- Fixed prompt file formatting to include trailing newlines at EOF for consistency across all prompt markdown files
 - Fixed `gh_pr_diff` to preserve raw patch content instead of normalizing tabs and whitespace
- Fixed `gh_pr_view` to include pull request reviews alongside issue-style comments for discussion context
- Fixed `gh_run_watch` and bundled `/green` to watch the workflow runs for the current `HEAD` commit instead of whichever branch run was newest
+- Fixed `gh_pr_view` to include inline review comments alongside pull request reviews and issue-style comments for discussion context
+- Fixed `gh_run_watch` to resolve explicit branch watches against the selected branch head instead of local `HEAD`
+- Fixed `gh_run_watch` to hide repo and polling internals from the tool schema and save full failed-job logs as session artifacts alongside the inline tailed output
+- Fixed `gh_*` tool outputs to spill full large results to artifacts instead of pre-truncating the head with unusable `offset=` guidance
+- Fixed bundled `/green` to watch the workflow runs for the current `HEAD` commit instead of whichever branch run was newest
 - Fixed OpenAI Responses session rehydration to strip stale assistant replay payloads before resumed requests ([#594](https://github.com/can1357/oh-my-pi/pull/594) by [@daandden](https://github.com/daandden))
 - Fixed inline image rendering to cap image height and preserve multiplexer scrollback during terminal resizes ([#587](https://github.com/can1357/oh-my-pi/pull/587) by [@smileynet](https://github.com/smileynet))

@@ -6497,4 +6508,4 @@ Initial public release.
 - Git branch display in footer
 - Message queueing during streaming responses
 - OAuth integration for Gmail and Google Calendar access
- HTML export with syntax highlighting and collapsible sections
+- HTML export with syntax highlighting and collapsible sections
@@ -10,7 +10,7 @@
 * 5. Compact markdown tables (remove padding)
 * 6. Collapse 2+ blank lines to single blank line
 * 7. Trim trailing whitespace (preserve indentation)
- * 8. No trailing newline at EOF
+ * 8. Trailing newline at EOF (disk only; runtime omits it)
 * 9. Bold RFC 2119 keywords (MUST, SHOULD, MAY, etc.) in prompt content
 */
 import { Glob } from "bun";
@@ -44,11 +44,11 @@ async function main() {
 		const original = await Bun.file(fullPath).text();
 		const formatted = formatPromptContent(original, PROMPT_FORMAT_OPTIONS);

-		if (original !== formatted) {
+		if (original !== `${formatted}\n`) {
 			if (check) {
 				console.log(`Would format: ${fullPath}`);
 			} else {
-				await Bun.write(fullPath, formatted);
+				await Bun.write(fullPath, `${formatted}\n`);
 				console.log(`Formatted: ${fullPath}`);
 			}
 			changed++;
@@ -19,4 +19,4 @@ Return concise JSON object with:
 Consider how file's changes relate to above files.
 {{/if}}

-Call submit_result tool with JSON payload.
+Call submit_result tool with JSON payload.
@@ -22,4 +22,4 @@ May include entries from list in propose_changelog `deletions` field for removal
 {{/each}}
 {{/if}}

-Use git_* tools to inspect changes. Call analyze_files for deeper per-file summaries. Finish with propose_commit or split_commit.
+Use git_* tools to inspect changes. Call analyze_files for deeper per-file summaries. Finish with propose_commit or split_commit.
@@ -1 +1 @@
-Split commit plan has {{count}} commits. Proceed? (y/N):
+Split commit plan has {{count}} commits. Proceed? (y/N):
@@ -35,4 +35,4 @@ Tool guidance:
 ## Changelog Requirements

 If changelog targets provided, you **MUST** call `propose_changelog` before finishing.
-If you propose split commit plan, include changelog target files in relevant commit changes.
+If you propose split commit plan, include changelog target files in relevant commit changes.
@@ -145,4 +145,4 @@ Call create_conventional_analysis with:
  "details": [],
  "issue_refs": []
 }
-</example>
+</example>
@@ -35,4 +35,4 @@
 {{/if}}
 <diff>
 {{ diff }}
-</diff>
+</diff>
@@ -47,4 +47,4 @@ Return ONLY valid JSON; no markdown fences or explanation.

 With entries: {"entries": {"Added": ["entry 1"], "Fixed": ["entry 2"]}}
 No changelog-worthy changes: {"entries": {}}
-</output-format>
+</output-format>
@@ -15,4 +15,4 @@ Already documented—skip these:

 <diff>
 {{ diff }}
-</diff>
+</diff>
@@ -21,4 +21,4 @@ Plain list, no preamble, no summary, no markdown formatting.
 - changed 'Connection::new()' to accept '&Config' instead of individual params
 </output-format>

-Observations only. Classification in reduce phase.
+Observations only. Classification in reduce phase.
@@ -5,4 +5,4 @@
 <related-files>
 {{ context_header }}
 </related-files>
-{{/if}}
+{{/if}}
@@ -47,4 +47,4 @@ Output:
 ],
 "issue_refs": []
 }
-</example>
+</example>
@@ -14,4 +14,4 @@

 <scope-candidates>
 {{ scope_candidates }}
-</scope-candidates>
+</scope-candidates>
@@ -1,3 +1,3 @@
 {{#if base_context}}
 {{ base_context }}
-{{/if}}Previous summary failed validation: {{ errors }}
+{{/if}}Previous summary failed validation: {{ errors }}
@@ -35,4 +35,4 @@ build | Updated serde to fix CVE-2024-1234
 </examples>
 <banned-words>
 comprehensive, various, several, improved, enhanced, quickly, simply, basically, this change, this commit, now
-</banned-words>
+</banned-words>
@@ -10,4 +10,4 @@

 <diff-stat>
 {{ stat }}
-</diff-stat>
+</diff-stat>
@@ -1,2 +1,2 @@
 Types: feat, fix, refactor, perf, docs, test, build, ci, chore, style, revert.
-Format: <type>(<scope>): <summary> with past-tense summary.
+Format: <type>(<scope>): <summary> with past-tense summary.
@@ -65,4 +65,4 @@ You **MAY** make file edits, create components, and run commands—and **SHOULD*
 Every interface should prompt "how was this made?" not "which AI made this?"
 You **MUST** commit to clear aesthetic direction and execute with precision.
 You **MUST** keep going until implementation is complete.
-</critical>
+</critical>
@@ -57,4 +57,4 @@ You **SHOULD** generally follow this procedure, but are allowed to adjust it as
 <critical>
 You **MUST** operate as read-only. You **MUST NOT** write, edit, or modify files, nor execute any state-changing commands, via git, build system, package manager, etc.
 You **MUST** keep going until complete.
-</critical>
+</critical>
@@ -7,4 +7,4 @@ description: {{jsonStringify description}}
 {{/if}}{{#if thinkingLevel}}thinking-level: {{jsonStringify thinkingLevel}}
 {{/if}}{{#if blocking}}blocking: true
 {{/if}}---
-{{body}}
+{{body}}
@@ -33,4 +33,4 @@ You will likely need to document these sections, but only take it as a starting

 <output>
 After analysis, you **MUST** write AGENTS.md to the project root.
-</output>
+</output>
@@ -118,4 +118,4 @@ Before acting, determine what kind of question this is:
 <critical>
 Source code is truth. Documentation is aspiration. Training data is history.
 You **MUST** keep going until you have a definitive, source-verified answer.
-</critical>
+</critical>
@@ -74,4 +74,4 @@ You **MUST NOT** pad with meta-commentary. Dense and useful beats long and thoro
 You **MUST** keep going until you have a clear answer or have exhausted available evidence.
 Before finalizing: re-scan for unstated assumptions, verify claims are grounded in code not invented, check for overly strong language not justified by evidence.
 This matters. The caller is stuck. Get it right.
-</critical>
+</critical>
@@ -46,4 +46,4 @@ You will likely need to document these sections, but only take it as a starting
 <critical>
 You **MUST** operate as read-only. You **MUST NOT** write, edit, or modify files, nor execute any state-changing commands, via git, build system, package manager, etc.
 You **MUST** keep going until complete.
-</critical>
+</critical>
@@ -124,4 +124,4 @@ Correctness ignores non-blocking issues (style, docs, nits).

 <critical>
 Every finding **MUST** be patch-anchored and evidence-backed.
-</critical>
+</critical>
@@ -13,4 +13,4 @@ You **MUST** maintain hyperfocus on the task at hand, do not deviate from what w
 - You **SHOULD** prefer edits to existing files over creating new ones.
 - You **MUST NOT** create documentation files (*.md) unless explicitly requested.
 - You **MUST** follow the assignment and the instructions given to you. You gave them for a reason.
-</directives>
+</directives>
@@ -33,4 +33,4 @@ Once CI is green, ensure the final commit is tagged `{{headTag}}` and push that
 <critical>
 The task is complete only when the workflow runs for the latest HEAD commit succeed.
 {{#if headTag}}The final green commit must be tagged `{{headTag}}` and that tag must be pushed.{{/if}}
-</critical>
+</critical>
@@ -2,4 +2,4 @@ The following is a summary of a branch that this conversation came back from:

 <summary>
 {{summary}}
-</summary>
+</summary>
@@ -1,2 +1,2 @@
 The user explored a different conversation branch before returning here.
-Summary of that exploration:
+Summary of that exploration:
@@ -27,4 +27,4 @@ You **MUST** use EXACT format:
 ## Next Steps
 1. [What should happen next to continue]

-Sections **MUST** be kept concise. You **MUST** preserve exact file paths, function names, error messages.
+Sections **MUST** be kept concise. You **MUST** preserve exact file paths, function names, error messages.
@@ -6,4 +6,4 @@ Rules:
 - **MUST NOT** mention running tests, builds, or other validation steps
 - **MUST NOT** explain what the user asked for
 - **MUST** write in first person (I added…, I fixed…)
- **MUST NOT** ask questions
+- **MUST NOT** ask questions
@@ -2,4 +2,4 @@ Another language model started to solve this problem and produced a summary of i

 <summary>
 {{summary}}
-</summary>
+</summary>
@@ -35,4 +35,4 @@ You **MUST** use this format (sections can be omitted if not applicable):

 You **MUST** output only the structured summary; you **MUST NOT** include extra text.

-Sections **MUST** be kept concise. You **MUST** preserve exact file paths, function names, error messages, and relevant tool outputs or command results. You **MUST** include repository state changes (branch, uncommitted changes) if mentioned.
+Sections **MUST** be kept concise. You **MUST** preserve exact file paths, function names, error messages, and relevant tool outputs or command results. You **MUST** include repository state changes (branch, uncommitted changes) if mentioned.
@@ -14,4 +14,4 @@ You **MUST** summarize the prefix to provide context for the retained suffix:

 You **MUST** output only the structured summary. You **MUST NOT** include extra text.

-You **MUST** be concise. You **MUST** preserve exact file paths, function names, error messages, and relevant tool outputs or command results if they appear. You **MUST** focus on what's needed to understand the kept suffix.
+You **MUST** be concise. You **MUST** preserve exact file paths, function names, error messages, and relevant tool outputs or command results if they appear. You **MUST** focus on what's needed to understand the kept suffix.
@@ -42,4 +42,4 @@ You **MUST** use this format (omit sections if not applicable):

 You **MUST** output only the structured summary; you **MUST NOT** include extra text.

-Sections **MUST** be kept concise. You **MUST** preserve relevant tool outputs/command results. You **MUST** include repository state changes (branch, uncommitted changes) if mentioned.
+Sections **MUST** be kept concise. You **MUST** preserve relevant tool outputs/command results. You **MUST** include repository state changes (branch, uncommitted changes) if mentioned.
@@ -27,4 +27,4 @@ Requirements:
 - scripts/templates/examples are optional. When present, each entry **MUST** write to skills/<name>/<bucket>/<path>.
 - You **MUST** only include files worth keeping long-term; you **MUST** omit stale assets so they are pruned.
 - You **MUST** preserve useful prior themes; you **MUST** remove stale or contradictory guidance.
- You **MUST** treat memory as advisory: current repository state wins.
+- You **MUST** treat memory as advisory: current repository state wins.
@@ -8,4 +8,4 @@ Operational rules:
 5) Conflict workflow: if memory disagrees with repo state or user instruction, you **MUST** prefer repo/user, treat memory as stale, proceed with corrected behavior, then update/regenerate memory artifacts through normal execution.
 6) You **MUST** escalate confidence only after repository verification; memory alone **MUST NOT** be treated as sufficient proof.
 Memory summary:
-{{memory_summary}}
+{{memory_summary}}
@@ -3,4 +3,4 @@ thread_id: {{thread_id}}
 Persistable response items (JSON):
 {{response_items_json}}

-You **MUST** extract durable memory now.
+You **MUST** extract durable memory now.
@@ -18,4 +18,4 @@ Rules:
 - rollout_summary: compact synopsis of what future runs should remember.
 - rollout_slug: short lowercase slug (letters/numbers/_), or null.
 - raw_memory: detailed durable memory blocks with enough context to reuse.
- If no durable signal exists, you **MUST** return empty strings for rollout_summary/raw_memory and null rollout_slug.
+- If no durable signal exists, you **MUST** return empty strings for rollout_summary/raw_memory and null rollout_slug.
@@ -61,4 +61,4 @@ _Full diff too large ({{len files}} files). Showing first ~{{linesPerFile}} line
 <diff>
 {{rawDiff}}
 </diff>
-{{/if}}
+{{/if}}
@@ -62,4 +62,4 @@ Key principles for your system prompts:
 - **MUST** make the agent proactive in seeking clarification when needed
 - **MUST** build in quality assurance and self-correction mechanisms

-The agents you create **MUST** be autonomous experts capable of handling their designated tasks with minimal additional guidance. Your system prompts are their complete operational manual.
+The agents you create **MUST** be autonomous experts capable of handling their designated tasks with minimal additional guidance. Your system prompts are their complete operational manual.
@@ -3,4 +3,4 @@ Design a custom agent for this request:
 {{request}}

 You **MUST** return only the JSON object required by your system instructions.
-You **MUST NOT** include markdown fences.
+You **MUST NOT** include markdown fences.
@@ -1 +1 @@
-Threshold-triggered maintenance: preserve critical implementation state and immediate next actions.
+Threshold-triggered maintenance: preserve critical implementation state and immediate next actions.
@@ -5,4 +5,4 @@ Do not use tools.
 Do not ask follow-up questions.
 Question:
 {{question}}
-</btw>
+</btw>
@@ -1,2 +1,2 @@
 Generate a concise git commit message from the provided diff. Use conventional commit format: `type(scope): description` where type is feat/fix/refactor/chore/test/docs and scope is optional. The description **MUST** be lowercase, imperative mood, no trailing period. Keep it under 72 characters.
-You **MUST** output ONLY the commit message, nothing else.
+You **MUST** output ONLY the commit message, nothing else.
@@ -60,4 +60,4 @@ You **MUST** read `rule://<name>` when working in that domain.
 </rules>
 {{/if}}
 Current date: {{date}}
-Current working directory: {{cwd}}
+Current working directory: {{cwd}}
@@ -10,4 +10,4 @@ You **MUST** keep exactly one task `in_progress` and all later tasks `pending`.

 After the initial `todo_write` call succeeds, continue with the user's request in the same turn.
 Do not emit another `todo_write` call unless task state materially changed.
-</system-reminder>
+</system-reminder>
@@ -7,4 +7,4 @@
 {{#xml "modified-files"}}
 {{join modifiedFiles "\n"}}
 {{/xml}}
-{{/if}}
+{{/if}}
@@ -43,4 +43,4 @@ Use exactly this structure:
 <instruction>
 Additional focus: {{additionalFocus}}
 </instruction>
-{{/if}}
+{{/if}}
@@ -104,4 +104,4 @@ Your turn ends ONLY by:

 You **MUST NOT** ask plan approval via text or `{{askToolName}}`; you **MUST** use `{{exitToolName}}`.
 You **MUST** keep going until complete.
-</critical>
+</critical>
@@ -20,4 +20,4 @@ If a `todo_write` call fails, you **MUST** fix the todo payload and retry before

 <critical>
 You **MUST** keep going until complete. This matters.
-</critical>
+</critical>
@@ -11,4 +11,4 @@ Plan file from previous session: `{{planFilePath}}`
 <instruction>
 If this plan is relevant to current work and not complete, you **MUST** continue executing it.
 If the plan is stale or unrelated, you **MUST** ignore it.
-</instruction>
+</instruction>
@@ -31,4 +31,4 @@ List 3-5 files most critical for implementing this plan:
 <critical>
 You **MUST** operate as read-only. You **MUST NOT** write, edit, or modify files, nor execute any state-changing commands, via git, build system, package manager, etc.
 You **MUST** keep going until complete.
-</critical>
+</critical>
@@ -6,4 +6,4 @@ You **MUST** choose exactly one next action now:
 2. Call `{{exitToolName}}` to finish planning and request approval

 You **MUST NOT** output plain text in this turn.
-</system-reminder>
+</system-reminder>
@@ -8,4 +8,4 @@ You **MUST** call submit_result as your only action now. Choose one:
 You **MUST NOT** give up if you can still complete the task through exploration (using available tools or repo context). If you submit an error, you **MUST** include what you tried and the exact blocker.

 You **MUST NOT** output text without a tool call. You **MUST** call submit_result to finish.
-</system-reminder>
+</system-reminder>
@@ -32,4 +32,4 @@ Your result **MUST** match this TypeScript interface:
 Giving up is a last resort. If truly blocked, you **MUST** call `submit_result` exactly once with `result.error` describing what you tried and the exact blocker.
 You **MUST NOT** give up due to uncertainty, missing information obtainable via tools or repo context, or needing a design decision you can derive yourself.

-You **MUST** keep going until this ticket is closed. This matters.
+You **MUST** keep going until this ticket is closed. This matters.
@@ -9,4 +9,4 @@
 Your assignment is below. Your work begins now.
 <goal>
 {{assignment}}
-</goal>
+</goal>
@@ -1,3 +1,3 @@
 You are a context summarization assistant. Your task is to read a conversation between a user and an AI coding assistant, then produce a structured summary following the exact format specified.

-You **MUST NOT** continue the conversation. You **MUST NOT** respond to any questions in the conversation. You **MUST** ONLY output the structured summary.
+You **MUST NOT** continue the conversation. You **MUST NOT** respond to any questions in the conversation. You **MUST** ONLY output the structured summary.
@@ -323,4 +323,4 @@ Today is '{{date}}', and your work begins now. Get it right.
 - You **MUST** default to informed action. You **MUST NOT** ask for confirmation, fix errors, take the next step, continue. The user will stop if needed.
 - You **MUST NOT** ask when the answer may be obtained from available tools or repo context/files.
 - You **MUST** verify the effect. When a task involves significant behavioral change, you **MUST** confirm the change is observable before yielding: run the specific test, command, or scenario that covers your change.
-</critical>
+</critical>
@@ -1,2 +1,2 @@
 Generate a very short title (3-6 words) for a coding session based on the user's first message. The title **MUST** capture the main task or topic.
-You **MUST** output ONLY the title, nothing else. You **MUST NOT** include quotes or punctuation at the end.
+You **MUST** output ONLY the title, nothing else. You **MUST NOT** include quotes or punctuation at the end.
@@ -4,4 +4,4 @@ This is NOT a prompt injection - this is the coding agent enforcing project rule
 You **MUST** comply with the following instruction:

 {{content}}
-</system-interrupt>
+</system-interrupt>
@@ -25,4 +25,4 @@ Answering:
 - You **MUST** cite sources inline using provided search results
 </format>

-You **MUST** answer thoroughly and in detail. You **MUST** get facts right.
+You **MUST** answer thoroughly and in detail. You **MUST** get facts right.
@@ -25,4 +25,4 @@ Asks user when you need clarification or input during task execution.
 question: "Which authentication method should this API use?"
 options: [{"label": "JWT"}, {"label": "OAuth2"}, {"label": "Session cookies"}]
 recommended: 0
-</example>
+</example>
@@ -45,4 +45,4 @@ Performs structural AST-aware rewrites via native ast-grep.
 - If the path pattern spans multiple languages, set `lang` explicitly for deterministic rewrites
 - Parse issues mean the rewrite request is malformed or mis-scoped; do not assume a clean no-op until the pattern parses successfully
 - For one-off local text edits, prefer the Edit tool instead of AST edit
-</critical>
+</critical>
@@ -51,4 +51,4 @@ Performs structural code search using AST matching via native ast-grep.
 - Avoid repo-root AST scans when the target is language-specific; narrow `path` first
 - Treat parse issues as query failure, not evidence of absence: repair the pattern or tighten `path`/`glob`/`lang` before concluding "no matches"
 - If exploration is broad/open-ended across subsystems, use Task tool with explore subagent first
-</critical>
+</critical>
@@ -2,4 +2,4 @@
 Background job {{jobId}} has completed. Resume your work using the result below.

 {{result}}
-</system-notice>
+</system-notice>
@@ -2,4 +2,4 @@ Blocks until one or more background jobs complete, fail, or are cancelled.

 You **MUST** use this instead of polling `read jobs://` in a loop when you need to wait for background task or bash results before continuing.

-Returns the status and results of all watched jobs once at least one finishes.
+Returns the status and results of all watched jobs once at least one finishes.
@@ -44,4 +44,4 @@ You **MUST** use specialized tools instead of bash for ALL file operations:
 - You **MUST NOT** use Bash for these operations like read, grep, find, edit, write, where specialized tools exist.
 - You **MUST NOT** use `2>&1` | `2>/dev/null` pattern, stdout and stderr are already merged.
 - You **MUST NOT** use `| head -n 50` or `| tail -n 100` pattern, use `head` and `tail` parameters instead.
-</critical>
+</critical>
@@ -21,4 +21,4 @@ Navigates, clicks, types, scrolls, drags, queries DOM content, and captures scre

 <output>
 Text for navigation/DOM queries, images for screenshots.
-</output>
+</output>
@@ -7,4 +7,4 @@ Performs basic calculations.

 <output>
 Returns each calculation result with its prefix and suffix applied.
-</output>
+</output>
@@ -2,4 +2,4 @@ Cancels a running background job started via async tool execution.

 You **SHOULD** use this when a background `bash` or `task` job is no longer needed or is stuck.

-You **MAY** inspect jobs first with `read jobs://` or `read jobs://<job-id>`.
+You **MAY** inspect jobs first with `read jobs://` or `read jobs://<job-id>`.
@@ -13,4 +13,4 @@ Typical flow:
 2. Perform exploratory work
 3. `rewind(report: …)` with concise findings

-After rewind, intermediate checkpoint messages are removed from active context and replaced by the report.
+After rewind, intermediate checkpoint messages are removed from active context and replaced by the report.
@@ -38,4 +38,4 @@ Unsure about auth method (OAuth vs JWT).

 <critical>
 You **MUST** only use when planning implementation steps. Research tasks (searching, reading, understanding) do not need this tool.
-</critical>
+</critical>
@@ -8,4 +8,4 @@ Retrieves content from a URL and returns it in a clean, readable format.

 <output>
 Returns processed, readable content. HTML transformed to remove boilerplate. PDF/DOCX converted to text. JSON returned formatted. With `raw: true`, returns untransformed HTML.
-</output>
+</output>
@@ -23,4 +23,4 @@ Matching file paths sorted by modification time (most recent first). Truncated a

 <avoid>
 For open-ended searches requiring multiple rounds of globbing and grepping, you **MUST** use Task tool instead.
-</avoid>
+</avoid>
@@ -4,4 +4,4 @@ Generates or edits images using Gemini image models.
 - You **MUST** provide a single detailed `subject` prompt for image generation or editing.
 - When using multiple `input`, you **SHOULD** describe each image's role directly in `subject`, e.g. `Image 1` for composition reference, `Image 2` for lighting reference, `Image 3` for background.
 - For text: you **SHOULD** add "sharp, legible, correctly spelled" for important text; keep text short
-</instructions>
+</instructions>
@@ -8,4 +8,4 @@ Reads a GitHub issue through the local GitHub CLI.

 <output>
 Returns issue metadata, body text, and optionally visible comments in a readable format.
-</output>
+</output>
@@ -9,4 +9,4 @@ Reads a GitHub pull request diff through the local GitHub CLI.

 <output>
 Returns a unified diff or changed-file list for the selected pull request.
-</output>
+</output>
@@ -3,9 +3,9 @@ Reads a GitHub pull request through the local GitHub CLI.
 <instruction>
 - Accepts a pull request number, URL, or branch name
 - Omitting `pr` targets the pull request associated with the current branch
- Use this for PR metadata, body, changed-file summaries, review discussion, and issue-comment context
+- Use this for PR metadata, body, changed-file summaries, review discussion, inline review comments, and issue-comment context
 </instruction>

 <output>
-Returns pull request metadata, body text, changed files, and optionally visible reviews and issue comments in a readable format.
-</output>
+Returns pull request metadata, body text, changed files, and optionally visible reviews, inline review comments, and issue comments in a readable format.
+</output>
@@ -8,4 +8,4 @@ Reads GitHub repository metadata using the local GitHub CLI.

 <output>
 Returns a concise repository summary including description, branch, visibility, stars, forks, and related metadata.
-</output>
+</output>
@@ -2,11 +2,11 @@ Watches a GitHub Actions workflow run through the local GitHub CLI.

 <instruction>
 - Accepts a run ID or full Actions run URL
- Omitting `run` watches the workflow runs for the current HEAD commit on the selected branch
+- Omitting `run` watches the workflow runs for the current HEAD commit in the current GitHub repository context
 - Omitting `branch` falls back to the current checked-out git branch
 - Fast-fails after the first detected job failure, waits briefly to collect concurrent failures, and then fetches tailed logs for the failed jobs
 </instruction>

 <output>
-Streams live run snapshots while polling, then returns the final run status, job list, and tailed logs for failed jobs when available. When `run` is omitted, the snapshots cover all workflow runs created for the current HEAD commit.
-</output>
+Streams live run snapshots while polling, then returns the final run status, job list, and tailed logs for failed jobs when available. When failed-job logs are fetched, the full failed-job logs are also saved as a session artifact for on-demand reads. When `run` is omitted, the snapshots cover all workflow runs created for the current HEAD commit.
+</output>
@@ -8,4 +8,4 @@ Searches GitHub issues through the local GitHub CLI.

 <output>
 Returns a concise list of matching issues with repository, state, labels, timestamps, and URLs.
-</output>
+</output>
@@ -8,4 +8,4 @@ Searches GitHub pull requests through the local GitHub CLI.

 <output>
 Returns a concise list of matching pull requests with repository, state, labels, timestamps, and URLs.
-</output>
+</output>
@@ -23,4 +23,4 @@ Searches files using powerful regex matching.
 - You **MUST** use Grep when searching for content.
 - You **MUST NOT** invoke `grep` or `rg` via Bash.
 - If the search is open-ended, requiring multiple rounds, you **MUST** use Task tool with explore subagent instead.
-</critical>
+</critical>
@@ -128,4 +128,4 @@ When adding a sibling declaration, prefer `prepend` on the next declaration.
 - For a range, either replace only the body or replace the whole range. Do not split range boundaries.
 - `content` must be literal file content with matching indentation. If the file uses tabs, use real tabs.
 - You **MUST NOT** use this tool to reformat or clean up unrelated code. **ALWAYS** use project-specific tooling like linters or code formatters which are much more efficient and reliable.
-  </critical>
+  </critical>
@@ -17,4 +17,4 @@ For OCR-style requests:

 For UI/screenshot debugging requests:
 - Focus on visible states, labels, toggles, error messages, disabled controls, and relevant affordances.
- Separate observed UI state from probable root cause.
+- Separate observed UI state from probable root cause.
@@ -29,4 +29,4 @@ Inspects an image file with a vision-capable model and returns compact text anal
 - Parameters are strict: only `path` and `question` are allowed
 - If image submission is blocked by settings, the tool will fail with an actionable error
 - If configured model does not support image input, configure a vision-capable model role before retrying
-</critical>
+</critical>
@@ -30,4 +30,4 @@ Interacts with Language Server Protocol servers for code intelligence.
 - Some operations require file to be saved to disk
 - Diagnostics glob mode samples up to 20 files per request to avoid long-running stalls on broad patterns
 - When `symbol` is provided for position-based actions, missing symbols or out-of-bounds `occurrence` values return an explicit error instead of silently falling back
-</caution>
+</caution>
@@ -69,4 +69,4 @@ edit {"path":"obsolete.txt","op":"delete"}
 - Generic anchors: `import`, `export`, `describe`, `function`, `const`
 - Repeating same addition in multiple hunks (duplicate blocks)
 - Full-file overwrites for minor changes (acceptable for major restructures or short files)
-</avoid>
+</avoid>
@@ -57,4 +57,4 @@ cells: [
    {"title": "use helper", "code": "configs = [parse_config(p) for p in Path('.').glob('*.json')]"}
 ]
 ```
-</example>
+</example>
@@ -25,4 +25,4 @@ Reads files from local filesystem or internal URLs.
 - You **MUST** use `read(path="dir/")` instead of `ls dir/` for directory listings.
 - You **MUST** always include the `path` parameter — NEVER call `read` with empty arguments `{}`.
 - When reading specific line ranges, use `offset` and `limit`: `read(path="file", offset=50, limit=100)` not `cat -n file | sed`.
-</critical>
+</critical>
@@ -6,4 +6,4 @@ Parameters:
 Behavior:
 - Returns ASCII diagram text.
 - Saves full ASCII output to an artifact URL (`artifact://<id>`) when artifact storage is available.
- Returns an error when the Mermaid input is invalid or rendering fails.
+- Returns an error when the Mermaid input is invalid or rendering fails.
@@ -33,4 +33,4 @@ For position-addressed or pattern-addressed changes, bash more efficient:

 Use Replace when _content itself_ identifies location.
 Use bash when _position_ or _pattern_ identifies what to change.
-</bash-alternatives>
+</bash-alternatives>
@@ -5,4 +5,4 @@ Resolves a pending preview action by either applying or discarding it.
 - `reason` is required and must explain why you chose to apply or discard.

 This tool is only valid when a pending action exists (typically after a preview step).
-If no pending action exists, the call fails with an error.
+If no pending action exists, the call fails with an error.
@@ -10,4 +10,4 @@ Requirements:

 Behavior:
 - If no checkpoint is active, this tool errors.
- On success, the session rewinds and keeps your report as retained context.
+- On success, the session rewinds and keeps your report as retained context.
@@ -31,4 +31,4 @@ Returns JSON with:
 - `query`
 - `activated_tools` — MCP tools activated by this search call
 - `match_count` — number of ranked matches returned by the search
- `total_tools`
+- `total_tools`
@@ -41,4 +41,4 @@ Command: `tasklist /v`
 Task: Get system info on "macbook"
 Host: macbook (10.0.0.20) | macos/zsh
 Command: `uname -a && sw_vers`
-</example>
+</example>
@@ -25,4 +25,4 @@
 {{mergeSummary}}
 </merge-summary>
 {{/if}}
-</task-summary>
+</task-summary>
@@ -128,4 +128,4 @@ Caller runs `bun check:ts` after both tasks complete. Tasks must NOT run it.
 ### Agent: {{name}}
 **Tools:** {{default (join tools ", ") "All"}}
 {{description}}
-{{/list}}
+{{/list}}
--- a/Show More
+++ b/Show More