docs: sync all markdown to v0.3.2 reality

[derek-palmer]

Summary

• SKILL.md (canonical + 2 copies): remove false "no CLI/MCP/hook/PyPI exists" claims; document what's live vs absent; fix config loader note; fix "planned, not yet runnable" on the CLI; update prompt paths to src/codeforerunner/prompts/; add audit/changelog to When To Activate; add generate/doctor/--fix/--stream to Workflow. V10 parity verified.
• AGENTS.md: updated prompt path, added generate/doctor commands, fixed _Rules tuple→dataclass note, updated drift-rule description.
• docs/getting-started.md: live PyPI install, generate/doctor --fix commands, fixed prompt paths, removed "PyPI not yet published" caveat.
• docs/editor-agent-setup.md: removed "no runtime wrapper exists yet".
• docs/prompt-guide.md: fixed prompt paths, added audit/changelog task rows.
• docs/roadmap.md: added P13 for v0.3.2 features.
• agent-configs/*.md: fixed prompt paths, removed "no CLI exists" caveats, added forerunner doc <task> usage.

Test plan

• scripts/validate_skill_copies.py — V10 parity OK
• pytest -q — 174 passed, 1 skipped
• forerunner check — no violations

🤖 Generated with Claude Code

Summary by CodeRabbit

• New Features

  • forerunner doctor --fix now generates starter configuration files
  • forerunner generate --stream enables token-by-token output streaming
  • New tasks available: security/dependency audits and changelog generation from git history
  • MCP server, pre-commit integration, and GitHub Actions workflow are now live

• Documentation

  • Comprehensive updates to AI agent setup and configuration guides
  • Clarified command flows and prompt assembly patterns

[coderabbitai]

Warning

Review limit reached

@derek-palmer, we couldn't start this review because you've used your available PR reviews for now.

Your plan currently allows 1 review/hour. Refill in 52 minutes and 4 seconds.

Your organization has run out of usage credits. Purchase more in the billing tab.

⌛ How to resolve this issue?

After more review capacity refills, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans have higher rate limits than trial, open-source, and free plans. In all cases, review capacity refills continuously over time.

Please see our FAQ for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 477c8afa-91a1-41fe-a78f-709c6769b554

📥 Commits

Reviewing files that changed from the base of the PR and between d68d0e0 and 35a7d74.

📒 Files selected for processing (5)

• agent-configs/cursor-rules.md
• agent/codeforerunner.skill.md
• docs/editor-agent-setup.md
• plugins/codeforerunner/skills/codeforerunner/SKILL.md
• skills/codeforerunner/SKILL.md

📝 Walkthrough

Walkthrough

This PR updates documentation across agent configurations, skill specifications, and product guides to reflect that the forerunner CLI ecosystem (CLI, MCP server, pre-commit hook, CI workflow, PyPI package) is now live. All instructions are unified to use command-based workflow (forerunner doc <task>) and prompt paths under src/codeforerunner/prompts/. New audit and changelog tasks are documented, and the P13 roadmap phase is marked complete.

Changes

CLI-Driven Agent Configuration and Skill Specification

Layer / File(s)	Summary
Agent-specific setup and configuration agent-configs/claude-project.md, agent-configs/cline.md, agent-configs/copilot-instructions.md, agent-configs/cursor-rules.md, agent-configs/windsurf.md, docs/editor-agent-setup.md	All agent configuration templates now direct setup through forerunner doc scan and task execution via forerunner doc <task>, with source prompts at src/codeforerunner/prompts/. Load order, common tasks, workflow steps, and repo boundary notes are updated consistently across editor platforms.
Codeforerunner skill specification agent/codeforerunner.skill.md, plugins/codeforerunner/skills/codeforerunner/SKILL.md, skills/codeforerunner/SKILL.md	Skill documents across three locations are updated to route agents via forerunner doc <task> or direct repo prompts, claim live CLI/MCP/hook/CI/PyPI tooling, add audit and changelog activation triggers, and revise context-layer and workflow instructions to use the command pattern. Safety rules now reference forerunner doctor --fix for config bootstrap instead of disclaiming runtime surfaces.

Internal Product Documentation and Structure

Layer / File(s)	Summary
Internal design and command reference AGENTS.md	Core prompt product location is documented as src/codeforerunner/prompts/, high-value commands expand to include forerunner generate --stream and forerunner doctor --fix, and drift constraints are broadened to include version-pin drift rules.
User-facing CLI guide and configuration docs/getting-started.md	Opening positioning emphasizes the broader product wrapper. "Use the CLI" section expands with complete command list (init, generate with streaming, doctor with fix, check, drift, mcp). Configuration section explains silent no-op when config is absent. "What Not To Do" is narrowed to remove Docker and Makefile assumptions.
Prompt composition and task definitions docs/prompt-guide.md	Guide clarifies three-layer sourcing from src/codeforerunner/prompts/ and assembly via forerunner doc <task>. Task prompt table is extended with audit.md (security/dependency audit) and changelog.md (Keep-a-Changelog generation from git history).
P13 phase completion and feature documentation docs/roadmap.md	New completed P13 phase lists inverse drift rules, version-pin drift detection (RV1), forerunner doctor --fix config bootstrap, forerunner generate --stream multi-provider streaming, and GitHub Action composite that installs and runs forerunner check with silent no-op when config is absent.

🎯 2 (Simple) | ⏱️ ~12 minutes

🐰 Prompts now flow through the CLI so clean,
Commands like doc <task> guide the machine,
Drift checks and audit tasks leap and spring,
Streaming and doctor bring new features' ring—
The ecosystem's alive, let the workflows sing! 🚀

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately and concisely summarizes the main change: synchronizing documentation to reflect v0.3.2 reality across all markdown files.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch docs/sync-to-v0.3.2

✨ Simplify code

• Create PR with simplified code
• Commit simplified code in branch docs/sync-to-v0.3.2

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

🧹 Nitpick comments (7)

agent-configs/cursor-rules.md (1)

7-8: ⚡ Quick win

Clarify the partial path reference.

Line 7 uses the full path src/codeforerunner/prompts/system/base.md, but line 8 references just "the context-format partial" without a file path or extension. For consistency and clarity, consider specifying partials/context-format.md or the full path src/codeforerunner/prompts/partials/context-format.md.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@agent-configs/cursor-rules.md` around lines 7 - 8, Update the ambiguous
partial reference so readers know the exact file: replace the phrase "the
`context-format` partial" with a clear path such as `partials/context-format.md`
or the full repository path
`src/codeforerunner/prompts/partials/context-format.md` to match the explicit
path used for `src/codeforerunner/prompts/system/base.md` and keep wording
consistent in the document.

plugins/codeforerunner/skills/codeforerunner/SKILL.md (1)

58-58: 💤 Low value

Optional: vary sentence structure for readability.

Three consecutive sentences begin with "To" which slightly impacts flow. Consider rewording for variety, for example:

• "Call a provider directly with forerunner generate <task> (add --stream for token-by-token output)."
• "Check drift using forerunner check."
• "Get a health report via forerunner doctor."

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@plugins/codeforerunner/skills/codeforerunner/SKILL.md` at line 58, Rewrite
the three consecutive sentences in SKILL.md to vary sentence openings for better
flow: replace "To call a provider directly: `forerunner generate <task>` (add
`--stream` for token-by-token output). To check drift: `forerunner check`. To
get a health report: `forerunner doctor`." with phrasing like "Call a provider
directly with `forerunner generate <task>` (add `--stream` for token-by-token
output). Check drift using `forerunner check`. Get a health report via
`forerunner doctor`." ensuring the commands remain exact (`forerunner generate
<task>`, `forerunner check`, `forerunner doctor`) and updating only the sentence
structure in SKILL.md.

agent-configs/claude-project.md (1)

7-9: 💤 Low value

Consider full paths for consistency.

Line 7 uses the full src/codeforerunner/prompts/system/base.md path, but line 8 uses the shorthand partials/context-format.md. For clarity, consider using the full path src/codeforerunner/prompts/partials/context-format.md to maintain consistency throughout the load order.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@agent-configs/claude-project.md` around lines 7 - 9, The load-order list
mixes a full path and a shorthand path; update the shorthand
"partials/context-format.md" to the full path
"src/codeforerunner/prompts/partials/context-format.md" so all references are
consistent (keep the existing items and wording otherwise) and verify the
assembled order still reads correctly in agent-configs/claude-project.md.

agent-configs/cline.md (1)

7-8: 💤 Low value

Consider full paths for consistency.

Line 7 uses the full src/codeforerunner/prompts/system/base.md path, but line 8 uses shorthand partials/output-rules.md and partials/context-format.md. For consistency, consider using full paths like src/codeforerunner/prompts/partials/output-rules.md and src/codeforerunner/prompts/partials/context-format.md.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@agent-configs/cline.md` around lines 7 - 8, The file uses a mix of full and
relative prompt paths which is inconsistent; update the shorthand paths
`partials/output-rules.md` and `partials/context-format.md` to their full
counterparts `src/codeforerunner/prompts/partials/output-rules.md` and
`src/codeforerunner/prompts/partials/context-format.md` so they match the
existing full path `src/codeforerunner/prompts/system/base.md`; locate the lines
referencing `partials/output-rules.md` and `partials/context-format.md` in
cline.md and replace them with the full
`src/codeforerunner/prompts/partials/...` paths.

agent/codeforerunner.skill.md (1)

58-58: 💤 Low value

Optional: vary sentence structure for readability.

Three consecutive sentences begin with "To" which slightly impacts flow. Consider rewording for variety, for example:

• "Call a provider directly with forerunner generate <task> (add --stream for token-by-token output)."
• "Check drift using forerunner check."
• "Get a health report via forerunner doctor."

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@agent/codeforerunner.skill.md` at line 58, The three consecutive sentences
all start with "To", which hurts readability; rephrase them to vary sentence
openings by using the provided alternatives and referencing the existing
commands: change "To call a provider directly: `forerunner generate <task>` (add
`--stream` for token-by-token output)." to something like "Call a provider
directly with `forerunner generate <task>` (add `--stream` for token-by-token
output).", change "To check drift: `forerunner check`." to "Check drift using
`forerunner check`.", and change "To get a health report: `forerunner doctor`."
to "Get a health report via `forerunner doctor`." ensuring the commands
`forerunner generate`, `forerunner check`, and `forerunner doctor` remain
unchanged.

docs/editor-agent-setup.md (1)

18-19: ⚡ Quick win

Clarify partial references.

Line 18 uses the full path src/codeforerunner/prompts/system/base.md, but line 19 references context-format and output-rules partials without the partials/ directory prefix or .md extensions. For consistency and clarity, consider using either partials/context-format.md and partials/output-rules.md, or the full paths src/codeforerunner/prompts/partials/context-format.md and src/codeforerunner/prompts/partials/output-rules.md.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@docs/editor-agent-setup.md` around lines 18 - 19, The docs reference the
assembled system prompt (base.md) but use inconsistent names for the partials;
update the third bullet so the partials are referenced explicitly and
consistently — e.g., change `context-format` and `output-rules` to
`partials/context-format.md` and `partials/output-rules.md` (or, if you prefer
the full source bundle style, use the corresponding full prompt filenames for
those partials) so the reader can unambiguously locate the partials when
assembling the agent context.

agent-configs/copilot-instructions.md (1)

7-8: 💤 Low value

Consider full paths for consistency.

Line 7 uses the full src/codeforerunner/prompts/system/base.md path, but line 8 uses shorthand partials/context-format.md. For consistency, consider the full path src/codeforerunner/prompts/partials/context-format.md.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@agent-configs/copilot-instructions.md` around lines 7 - 8, Replace the
shorthand reference "partials/context-format.md" with the full, consistent path
"src/codeforerunner/prompts/partials/context-format.md" in the
agent-configs/copilot-instructions.md entry so both lines use full paths like
"src/codeforerunner/prompts/system/base.md" and
"src/codeforerunner/prompts/partials/context-format.md".

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@agent-configs/claude-project.md`:
- Around line 7-9: The load-order list mixes a full path and a shorthand path;
update the shorthand "partials/context-format.md" to the full path
"src/codeforerunner/prompts/partials/context-format.md" so all references are
consistent (keep the existing items and wording otherwise) and verify the
assembled order still reads correctly in agent-configs/claude-project.md.

In `@agent-configs/cline.md`:
- Around line 7-8: The file uses a mix of full and relative prompt paths which
is inconsistent; update the shorthand paths `partials/output-rules.md` and
`partials/context-format.md` to their full counterparts
`src/codeforerunner/prompts/partials/output-rules.md` and
`src/codeforerunner/prompts/partials/context-format.md` so they match the
existing full path `src/codeforerunner/prompts/system/base.md`; locate the lines
referencing `partials/output-rules.md` and `partials/context-format.md` in
cline.md and replace them with the full
`src/codeforerunner/prompts/partials/...` paths.

In `@agent-configs/copilot-instructions.md`:
- Around line 7-8: Replace the shorthand reference "partials/context-format.md"
with the full, consistent path
"src/codeforerunner/prompts/partials/context-format.md" in the
agent-configs/copilot-instructions.md entry so both lines use full paths like
"src/codeforerunner/prompts/system/base.md" and
"src/codeforerunner/prompts/partials/context-format.md".

In `@agent-configs/cursor-rules.md`:
- Around line 7-8: Update the ambiguous partial reference so readers know the
exact file: replace the phrase "the `context-format` partial" with a clear path
such as `partials/context-format.md` or the full repository path
`src/codeforerunner/prompts/partials/context-format.md` to match the explicit
path used for `src/codeforerunner/prompts/system/base.md` and keep wording
consistent in the document.

In `@agent/codeforerunner.skill.md`:
- Line 58: The three consecutive sentences all start with "To", which hurts
readability; rephrase them to vary sentence openings by using the provided
alternatives and referencing the existing commands: change "To call a provider
directly: `forerunner generate <task>` (add `--stream` for token-by-token
output)." to something like "Call a provider directly with `forerunner generate
<task>` (add `--stream` for token-by-token output).", change "To check drift:
`forerunner check`." to "Check drift using `forerunner check`.", and change "To
get a health report: `forerunner doctor`." to "Get a health report via
`forerunner doctor`." ensuring the commands `forerunner generate`, `forerunner
check`, and `forerunner doctor` remain unchanged.

In `@docs/editor-agent-setup.md`:
- Around line 18-19: The docs reference the assembled system prompt (base.md)
but use inconsistent names for the partials; update the third bullet so the
partials are referenced explicitly and consistently — e.g., change
`context-format` and `output-rules` to `partials/context-format.md` and
`partials/output-rules.md` (or, if you prefer the full source bundle style, use
the corresponding full prompt filenames for those partials) so the reader can
unambiguously locate the partials when assembling the agent context.

In `@plugins/codeforerunner/skills/codeforerunner/SKILL.md`:
- Line 58: Rewrite the three consecutive sentences in SKILL.md to vary sentence
openings for better flow: replace "To call a provider directly: `forerunner
generate <task>` (add `--stream` for token-by-token output). To check drift:
`forerunner check`. To get a health report: `forerunner doctor`." with phrasing
like "Call a provider directly with `forerunner generate <task>` (add `--stream`
for token-by-token output). Check drift using `forerunner check`. Get a health
report via `forerunner doctor`." ensuring the commands remain exact (`forerunner
generate <task>`, `forerunner check`, `forerunner doctor`) and updating only the
sentence structure in SKILL.md.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: c3809a65-061c-49df-bf10-a0ca2daef584

📥 Commits

Reviewing files that changed from the base of the PR and between 4744b17 and d68d0e0.

📒 Files selected for processing (13)

• AGENTS.md
• agent-configs/claude-project.md
• agent-configs/cline.md
• agent-configs/copilot-instructions.md
• agent-configs/cursor-rules.md
• agent-configs/windsurf.md
• agent/codeforerunner.skill.md
• docs/editor-agent-setup.md
• docs/getting-started.md
• docs/prompt-guide.md
• docs/roadmap.md
• plugins/codeforerunner/skills/codeforerunner/SKILL.md
• skills/codeforerunner/SKILL.md