task-harness
Initialize a durable task pack under harness/tasks so long-running work has stable brief, progress, decisions, evidence, and verdict artifacts.
Skill by gclm
Chalk Labs
Need help deploying task-harness?
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/gclm-task-harness
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
Use cases: productivity
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. Checked Oct 25, 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/task-harness
What's inside
- Skill name
task-harness- Description
- Initialize a durable task pack under harness/tasks so long-running work has stable brief, progress, decisions, evidence, and verdict artifacts.
- License
- MIT
- Instructions
- 28 lines in SKILL.md · view on GitHub
Files (9)
- SKILL.md
- evidence-index.json · assets/
- harness-state.md · assets/
- task-brief.md · assets/
- task-decisions.md · assets/
- task-progress.md · assets/
- verdict.json · assets/
- progress-and-checkpoint-rules.md · references/
- task-artifact-contract.md · references/
SKILL.md (first lines)
# Task Harness Use when the work should be anchored in `harness/tasks/<task-id>/` instead of only living in chat context. ## Read First - `skills/harness/task-harness/references/task-artifact-contract.md` - `skills/harness/task-harness/references/progress-and-checkpoint-rules.md` - `skills/harness/task-harness/assets/task-brief.md` - `skills/harness/task-harness/assets/task-progress.md` - `skills/harness/task-harness/assets/task-decisions.md` - `skills/harness/task-harness/assets/evidence-index.json` - `skills/harness/task-harness/assets/verdict.json` ## Goals - initialize a durable task pack - keep task scope explicit - leave stable artifacts for later `review / verify / qa` ## Process 1. choose or create a `task-id` 2. initialize `harness/tasks/<task-id>/` 3. write the first brief and progress snapshot 4. route the next step to `build`, `review`, `verify`, or `qa`
From gclm/super-stack, MIT.
More from gclm/super-stack
- 0.414score
doc
gclm
Read, create, or edit `.docx` documents.
SkillDFreshness: DClaude Code +9 - 0.414score
pdf
gclm
Read, create, or review PDF files.
SkillDFreshness: DClaude Code +9 - 0.414score
low-altitude-interviewer
gclm
Screen resumes and prepare interviews for low-altitude aviation roles such as drone inspection, facade cleaning, and related operational backends.
SkillDFreshness: DClaude Code +9 - 0.411score
closed-loop-testing
gclm
Collect fresh evidence into a task-local evidence pack and turn implemented work into a structured verdict before calling it done.
SkillDFreshness: DClaude Code +9 - 0.411score
frontend-design
gclm
Design or reshape frontend UI with an explicit visual direction, stronger aesthetic judgment, and browser-verifiable outcomes instead of generic AI-looking…
SkillDFreshness: DClaude Code +9 - 0.411score
frontend-refactor
gclm
Refactor frontend code by separating structural cleanup, UI consistency, state flow, and verification instead of mixing visual edits with risky behavior…
SkillDFreshness: DClaude Code +9 - 0.411score
harness-marathon
gclm
Run long tasks with checkpoints, resume discipline, and human escalation gates so work can continue across turns without losing context.
SkillDFreshness: DClaude Code +9 - 0.411score
brainstorm
gclm
Explore 2-3 credible implementation approaches, compare trade-offs, and converge on a recommendation before detailed planning.
SkillDFreshness: DClaude Code +9 - 0.411score
repo-bootstrap
gclm
Inspect a repository and decide whether to initialize the generated-project scaffold before task-level harness work begins.
SkillDFreshness: DClaude Code +9 - 0.411score
database-design
gclm
Design or review database schemas, indexes, constraints, and data access trade-offs before implementation hardens around weak persistence choices.
SkillDFreshness: DClaude Code +9 - 0.411score
codex-record-retrospective
gclm
Review Codex local records for a specific project path, reconstruct what happened, and generate reusable retrospective or skill-evolution recommendations for…
SkillDFreshness: DClaude Code +9 - 0.411score
architecture-guardrails
gclm
Keep long-running work inside agreed boundaries by tracking scope drift, upgrade conditions, and design decisions before implementation silently expands.
SkillDFreshness: DClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Mar 29, 2026 (7 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/gclm/super-stack
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from gclm
- 0.414score
pdf
gclm
Read, create, or review PDF files.
SkillDFreshness: DClaude Code +9 - 0.414score
low-altitude-interviewer
gclm
Screen resumes and prepare interviews for low-altitude aviation roles such as drone inspection, facade cleaning, and related operational backends.
SkillDFreshness: DClaude Code +9 - 0.414score
doc
gclm
Read, create, or edit `.docx` documents.
SkillDFreshness: DClaude Code +9 - 0.411score
frontend-refactor
gclm
Refactor frontend code by separating structural cleanup, UI consistency, state flow, and verification instead of mixing visual edits with risky behavior…
SkillDFreshness: DClaude Code +9 - 0.411score
frontend-design
gclm
Design or reshape frontend UI with an explicit visual direction, stronger aesthetic judgment, and browser-verifiable outcomes instead of generic AI-looking…
SkillDFreshness: DClaude Code +9 - 0.411score
repo-bootstrap
gclm
Inspect a repository and decide whether to initialize the generated-project scaffold before task-level harness work begins.
SkillDFreshness: DClaude 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 task-harness 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
task-harness skill. Agent Mart. https://agentmart.dev/skills/gclm-task-harness (last updated 2026-10-24)