forge-doctor
Skill2 starsChanged 5 months agoMITView on GitHub
Run the Forge diagnostic report to surface environment, plugin, project,
2 other projects use a version of this.
Adopted by
Most recently taken up by
- ForgeyClap/claude-forgetheir own version, first seen Oct 11, 2026
- ForgeyClap/Forge-Coretheir own version, first seen Oct 11, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- ForgeyClap/claude-forge★ 2same name
- ForgeyClap/Forge-Core★ 2same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository46 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- forge-opencode/CLAUDE.mdCLAUDE.md
- forge-adoptSkill
- forge-arch-proSkill
- forge-archSkill
- forge-autopilotSkill
- forge-autopilot-stopSkill
- forge-build-proSkill
- forge-buildSkill
- forge-deploySkill
- dreamer-runSkill
- forge-evalSkill
- forge-feedbackSkill
- flowSkill
- forge-force-advanceSkill
- graduateSkill
- health-checkSkill
- forge-initSkill
- forge-monitorSkill
- forge-orchestrateSkill
- forge-plan-proSkill
- forge-planSkill
- preflightSkill
- forge-product-proSkill
- forge-productSkill
- forge-releaseSkill
- forge-resolveSkill
- forge-resumeSkill
- forge-retroSkill
- forge-reviewSkill
- forge-rulesSkill
- forge-set-profileSkill
- skill-creatorSkill
- forge-spec-proSkill
- forge-specSkill
- forge-sprint-proSkill
- forge-sprintSkill
- forge-srs-proSkill
- forge-srsSkill
- forge-statusSkill
- forge-trace-matrixSkill
- forge-uninstallSkill
- forge-validateSkill
- watchSkill
- watch-stopSkill
- forge-whySkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 31
If $ARGUMENTS contains a path, use it
- Distinctive: line 33
Run the diagnostic. Default invocation:
- Very specific: line 37
If the user wants only failures and warnings (no passes), add --quiet
- Very specific: line 38
If the user wants machine-readable output (parsing in a script), add --json
- Very specific: line 41
If the report shows recent entries under Recent errors, recommend the user view .forge/hook-errors.log for full tracebacks
- Distinctive: line 43
After presenting the report, only restate the recommended next steps if the user explicitly asks "what should I do?"
- Distinctive: line 48
After running, all of these should be true:
- Distinctive: line 52
If any check failed, the user knows the exact command to run next (verbatim from output)
- Very specific: line 53
If the hook_errors check warned, the user has been pointed at .forge/hook-errors.log
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Forge Doctor
Runs scripts/doctor.py, which performs 13 deterministic checks across environment,
plugin integrity, project state, and global state. Each failing check includes a specific
fix command. The script exits 0 if all checks pass (warnings allowed), 1 if any fail.
When to Use
- User types
/forge:doctoror/forge:health - User says "Forge isn't working", "hooks aren't firing", "something's wrong with Forge"
- User reports a Forge bug — run this BEFORE suggesting reinstall or workarounds
- Pre-flight check before a Forge upgrade or before tagging a release
- User asks "why isn't
/forge:initrunning?" or any other Forge command-failure question - User pastes an error message that mentions Forge, hooks, or
pipeline/
Steps
- Determine the project directory:
- If
$ARGUMENTScontains a path, use it - Otherwise use
${CLAUDE_PROJECT_DIR}if set, else the current working directory
- If
- Run the diagnostic. Default invocation:
python ${CLAUDE_PLUGIN_ROOT}/scripts/doctor.py --cwd "<project>" - If the user wants only failures and warnings (no passes), add
--quiet - If the user wants machine-readable output (parsing in a script), add
--json - Present the output to the user verbatim. Do NOT paraphrase the
Fix:lines — they are literal shell commands the user must run, not English advice - If the report shows recent entries under
Recent errors, recommend the user view.forge/hook-errors.logfor full tracebacks - After presenting the report, only restate the recommended next steps if the user explicitly asks "what should I do?"
Verification
After running, all of these should be true:
- The full diagnostic report is shown to the user, including any per-check
Fix:lines - The exit code has been interpreted: 0 means healthy (warnings allowed), 1 means at least one hard failure
- If any check failed, the user knows the exact command to run next (verbatim from output)
- If the
hook_errorscheck warned, the user has been pointed at.forge/hook-errors.log
Examples
User: "Forge isn't injecting anything at session start"
Claude: [runs python ${CLAUDE_PLUGIN_ROOT}/scripts/doctor.py --cwd "$(pwd)"]
Forge Doctor — diagnostic report
=================================
Environment:
✓ Python 3.11.7
✓ PyYAML 6.0.1
✗ Claude Code 2.0.4 (≥ 2.1.0 required)
Fix: Upgrade Claude Code via your installer
Plugin:
✓ plugin.json valid (name=forge, version=0.1.3)
✓ All 7 hooks present
✓ 16 agent definitions (≥ 12 expected)
✓ Gate criteria parsed: 60 criteria across 12 stages
Result: 1 failure."Your Claude Code is below the minimum required version — SessionStart hooks rely on the
2.1.0 schema, so they're being ignored. Upgrade Claude Code via your installer, then run
/forge:doctor again to confirm."
User: "/forge:doctor --json"
Claude: [runs with --json, presents JSON output verbatim without paraphrasing]
User: "Just tell me what's broken with Forge"
Claude: [runs with --quiet --cwd "$(pwd)", presents only failures/warnings]
"Two things need attention:
- ⚠
.forge/not in.gitignore(runtime state shouldn't be committed) Fix:echo '.forge/' >> .gitignore - ⚠ Last hook errors:
{"ts": "2026-05-15T10:23Z", "hook": "stop-reflect", "kind": "timeout", ...}
Fix: Investigate
/home/user/proj/.forge/hook-errors.log"
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

