/forge-validate
Run validation on a specific module from the current forge plan
$ npx -y skills add TT-Wang/forge --skill forge-validate --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/forge-validate
Context preview
The summary Claude sees to decide when to auto-load this skill.
Run validation on a specific module from the current forge plan
SKILL.md
forge-validate.SKILL.mdname: forge-validate description: Run validation on a specific module from the current forge plan argument-hint: <module-id> allowed-tools: [Read, Bash, mcp__forge__validate, mcp__forge__iteration_state]
Validate a specific module from the active forge plan. Prefix all output with `[forge:validate]`.
1. Read the latest plan from `.forge/plans/` (most recent JSON file) 2. Find the module matching the argument (e.g., "m1", "m2") 3. Call mcp__forge__validate with that module's verify commands and files 4. Call mcp__forge__iteration_state to get retry history 5. Report results:
## Validation: {module title}
**Status:** PASSED / FAILED
**Score:** {X}/{Y} checks passed
**Stagnant:** yes/no
**Attempt:** {N}
### Results
- [PASS] npm test -- --grep 'auth'
- [FAIL] npx tsc --noEmit (error: TS2345 in src/auth.ts:42)
### Recommendation
{PASS / RETRY_WITH_DEBUGGER / ESCALATE}Turn Claude Code into a structured delivery loop: plan the work, run modules in parallel, validate deeply, retry intelligently, and carry forward what worked.
Repo: TT-Wang/forge

