Chalk Labs
Need help deploying st-inspect?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/hyusecs-st-inspect
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/st-inspect
What's inside
- Skill name
st-inspect- Description
- Full-lane Inspect phase. Run st-inspect (cross-artifact analysis and validate checks in one pass), bring decisions to the user, and stop at the gate "user OKs the build".
- License
- MIT
- Instructions
- 38 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# /stratum:st-inspect You are the orchestrator. You route and verify. Plugin root (`<plugin root>`) is two levels up from this skill's base directory. ## 0. Start 1. Read `<project>/.stratum/state.json`. If there is no `feature_directory`, stop and tell the user to run `/stratum:st-define`. ## Steps 1. Note `git status --porcelain` and `git diff -- <feature>/plan.md`. Start the `st-inspect` subagent (`stratum:st-inspect`) with absolute paths: plugin root, project root, feature dir. Tell it: follow `<plugin root>/procedures/analyze.md` (`<PLUGIN_ROOT>` is the plugin root), then its validate checks. After it returns, run `git status --porcelain` again: only `plan.md` may change. Run `git diff -- <feature>/plan.md` again and compare it with the noted diff: it touches only its `## Validate` section. Any other change is a finding. 2. **Findings.** Check each finding against the source and drop the false ones (see `/stratum:st`). Big decisions go to the user, one at a time, each with your recommendation. Small ones you fix and list at the gate. Fixes to spec, plan or tasks go to the `st-plan` subagent (resume it with `SendMessage` if it is still alive); have `st-git` commit them. 3. If any CRITICAL finding was fixed, run `st-inspect` once more. It replaces its older `## Validate` section. 4. If a fix removed or renamed a task in step 2, grep each old ID from the fix's diff of `tasks.md` in the `## Validate` section of `plan.md`; on any hit run `st-inspect` once more. ## Gate: user OKs the build Run `bash <plugin root>/scripts/st-advisor.sh check plan <feature dir>` and show its line; `advisor approval is stale` counts as no approval: run the plan point (`st-plan` step 4) once more, as a round, before the gate. If the plan changed at Inspect only by fixes of findings, with no approach change, run `bash <plugin root>/scripts/st-advisor.sh waive plan <feature dir> "<reason>"` in place of the plan point. Skip the waive, and `check plan`, when the plan point printed `advisors: none` or a `no advisor` line. Have `st-git` commit `<feature dir>/advisor.md` if it changed. Show in short form: coverage, remaining findings by severity, small fixes made, findings dropped, and the verdict. At "OK to build?", show the `WAIVED` line and its reason. With no advisor at the plan point, say so in place of the `WAIVED` line. Ask: "OK to build?" Start `/stratum:st-build` only on yes. ## Always - Never skip the commit guard. If a commit is blocked, tell the user; do not work around it. - Push only when the user says "push".
From HyuseCS/stratum, MIT.
More from HyuseCS/stratum
- 0.604score
st-handoff
HyuseCS
Write the session handoff to .stratum/handoff.md so the next session starts where this one stopped.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
st-ponytail-audit
HyuseCS
Quality audit of a whole repo: bugs, security holes, what breaks under real load, risky code without tests, slow paths, and what to delete, merge or split.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-build
HyuseCS
Full-lane Build phase.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-define
HyuseCS
Full-lane Define phase.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-model
HyuseCS
Set the model or effort a Stratum agent uses in this project, saved in .stratum/models.json.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-plan
HyuseCS
Full-lane Plan phase.
SkillAFreshness: A★ 3Claude Code +9 - 0.600score
st-graphify
HyuseCS
Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question…
SkillAFreshness: A★ 3Claude Code +9 - 0.597score
st-close
HyuseCS
Full-lane Close phase.
SkillAFreshness: A★ 3Claude Code +9 - 0.596score
st-issues
HyuseCS
Turn the current feature's tasks.md into GitHub issues on the repo's own GitHub remote, skipping tasks that already have one.
SkillAFreshness: A★ 3Claude Code +9 - 0.596score
st-git
HyuseCS
Commit on demand through the st-git agent.
SkillAFreshness: A★ 3Claude Code +9 - 0.596score
st-debug
HyuseCS
Run the st-debug agent on demand for one task of the current feature, then verify with st-verify.sh (test green, no added comment lines).
SkillAFreshness: A★ 3Claude Code +9 - 0.596score
st-constitution
HyuseCS
Create or amend the project rules in .stratum/constitution.md from the user's words, with a version bump and sync report.
SkillAFreshness: A★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 0
- Open issues
- 4
- Last commit (this folder)
- Oct 23, 2026 (9 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/HyuseCS/stratum
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from HyuseCS
- 0.604score
st-handoff
HyuseCS
Write the session handoff to .stratum/handoff.md so the next session starts where this one stopped.
SkillAFreshness: A★ 3Claude Code +9 - 0.604score
st-ponytail-audit
HyuseCS
Quality audit of a whole repo: bugs, security holes, what breaks under real load, risky code without tests, slow paths, and what to delete, merge or split.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-plan
HyuseCS
Full-lane Plan phase.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-define
HyuseCS
Full-lane Define phase.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-build
HyuseCS
Full-lane Build phase.
SkillAFreshness: A★ 3Claude Code +9 - 0.601score
st-model
HyuseCS
Set the model or effort a Stratum agent uses in this project, saved in .stratum/models.json.
SkillAFreshness: A★ 3Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get st-inspect running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
st-inspect skill. Agent Mart. https://agentmart.dev/skills/hyusecs-st-inspect (last updated 2026-10-24)