codagent-ai/agent-validator/tasks
Claude Code and agent skill by codagent-ai/agent-validator
- community
Checking that the SKILL.md file is still in the repository.
Install
npx vskill@latest install codagent-ai/agent-validator/tasksSKILL.md
Task: Skip CLI command and skill
Goal
Add an agent-gauntlet skip CLI subcommand and a /gauntlet-skip skill that advance the gauntlet execution state baseline to the current working tree without running any verification gates. This lets users acknowledge current changes so the next agent-gauntlet run only diffs against changes made after the skip.
Background
You MUST read these files before starting:
design.mdfor the full design rationale and decisionsspecs/skip-command/spec.mdfor all acceptance criteria
Gauntlet tracks a baseline in gauntlet_logs/.execution_state (branch, commit SHA, working tree ref via git stash create). Each run diffs against this baseline. The skip command advances this baseline without running gates — same end state as a successful run.
Key files to modify:
src/commands/skip.ts— create new file, following the pattern insrc/commands/clean.tssrc/commands/index.ts— addregisterSkipCommandexportsrc/index.ts— import and callregisterSkipCommand(program)skills/gauntlet-skip/SKILL.md— create new skill file, following the pattern inskills/gauntlet-status/SKILL.md
Key files to understand (read-only):
src/commands/clean.ts— the closest existing command pattern to follow (config loading, debug logger init, lock acquisition,cleanLogs()call)src/utils/execution-state.ts—writeExecutionState(logDir)writes.execution_statewith current branch/commit/working-tree-ref and preservesunhealthy_adapters;getCurrentCommit()returns the full SHAsrc/commands/shared.ts—cleanLogs(logDir, maxPreviousLogs)archives log files;acquireLock(logDir)/releaseLock(logDir)prevent concurrent execution
Implementation details:
- The
skip.tscommand handler should: load config, init debug logger, acquire lock, callcleanLogs(), callwriteExecutionState(), print confirmation with abbreviated commit SHA (first 7 chars ofgetCurrentCommit()), release lock, exit 0. - On error: release lock if acquired, print error message, exit 1.
- No flags or options — the command always snapshots "now".
- The skill SKILL.md should be minimal: run
agent-gauntlet skipand report the output. Setdisable-model-invocation: trueandallowed-tools: Bash.
Done When
agent-gauntlet skipcreates the log directory and writes.execution_statewhen no prior state exists (exit 0)agent-gauntlet skiparchives existing logs and overwrites.execution_statewhen prior state and logs exist (exit 0)agent-gauntlet skippreservesunhealthy_adaptersentries from existing.execution_stateagent-gauntlet skipprints a confirmation message including the abbreviated commit SHAagent-gauntlet skipexits non-zero with an error message when another gauntlet process holds the lock, without modifying state or logsagent-gauntlet skipexits non-zero with an error message when run outside a git repository/gauntlet-skipskill executesagent-gauntlet skipand reports the command output
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingopenspec/changes/archive/2026-02-23-gauntlet-skip/tasks/skip-command-and-skill.md
Popularity
- Repository stars
- 29
- Repository forks
- 2
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/codagent-ai/agent-validator/tasks)