agentleFS
Sign inStart free
sayeediftekharsayeediftekhar/lean-context/skills/harness-engineering/SKILL.md

harness-engineering

Skill0 starsChanged 4 months agoMITView on GitHub

Design or debug the harness around an AI coding agent — the steps, tools, loops, and gates that decide whether a capable model ships correct work. Use when planning how to structure a task safely; when designing a feedback loop or test that must go red on the real problem; when deciding which model a task needs (model routing); when setting up gates, sign-offs, or a commit/approval boundary; when choosing what tools/permissions an agent should have (least privilege); when setting up a session or handoff ritual; or when debugging a silent, false-green, or "agent did the wrong KIND of thing" failure. Keywords: feedback loop, verify-fact, model routing, gate, least privilege, prompt injection, false green, config-check theater, silent failure, four pillars.

25 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use self-improvement-loops, bdi-mental-states and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • DVwebBiz11Skill

    self-improvement-loops

    This skill should be used when the harness, scaffold, workflow, or optimizer itself is the optimization target: recursive self-improvement (RSI) loops, meta-harnesses, self-improving harnesses that mine their own failures and propose bounded edits, evolutionary or population-based search over agent scaffolds, acceptance gates for self-modifying systems, and agentic context evolution where the mechanism that produces context is versioned and evolved. Route governance of a single autonomous loop (locked surfaces, durable logs, rollback, novelty gates, approval boundaries) to harness-engineering, measurement and quality-gate design to evaluation, judge design to advanced-evaluation, and remote sandbox infrastructure to hosted-agents.

    Also in 4 projects that use this

  • Downey6689Skill

    bdi-mental-states

    Transforms external RDF context into BDI mental states. Invoke when modeling beliefs, desires, and intentions.

    Also in 5 projects that use this

  • Downey6689Skill

    hosted-agents

    Guidance for hosted/sandboxed agents. Invoke when running background agents in isolated VMs or multi-client setups.

    Also in 5 projects that use this

  • DVwebBiz11Skill

    long-horizon-prompting

    This skill should be used when writing, enhancing, or evaluating the launch prompt for a long-running autonomous agent or a parallel multi-agent orchestration attacking a hard problem: pseudo-formal task briefs that define terms and an exact success predicate linguistically, enumerate non-counting outcomes, set persistence rules with explicit stop and return conditions and effort floors, manage a diverse portfolio of parallel approaches with an approach registry and blocked-route bookkeeping, and gate the return on adversarial audit. Route agent topology and coordination protocols to multi-agent-patterns, runtime control surfaces and loop governance to harness-engineering, evaluator and quality-gate construction to evaluation, judge design to advanced-evaluation, and compaction or memory mechanics to context-compression and memory-systems.

    Also in 4 projects that use this

  • AbdelrhmanAlquranSkill

    latent-briefing

    This skill should be used when the user asks to "share memory between agents", "KV cache compaction for multi-agent", "orchestrator worker context", "latent briefing", "reduce worker tokens", "cross-agent memory without summarization", or discusses Attention Matching compaction, recursive language models with workers, or token explosion in hierarchical agents.

    Also in 4 projects that use this

  • abdullah1854Skill

    context-degradation

    Also in 5 projects that use this

  • +2 more from Downey6689

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 22 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read the relevant HARNESS.md section now, then design against it.

    line 62

“Common” is measured: kept by at least 6 in 10 of the 22 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Harness Engineering — the four pillars

This skill runs the harness half of this repo. The full reference is ${CLAUDE_PLUGIN_ROOT}/HARNESS.md; read it for the section the task needs.

First move: diagnose which half is failing

  • Agent did the wrong kind of action (skipped a step, self-committed, reached prod) → harness failure → this skill.
  • Agent did the right kind of action on wrong information → context failure → use the lean-context skill / PLAYBOOK.md instead.

The four pillars — every task needs all four

① STEPS   sequence that makes errors cheap to catch  (slow front, fast back)
② TOOLS   least privilege — worst mistake stays survivable
③ LOOPS   automated checks that go red on the ACTUAL problem
④ GATES   sign-offs that can't be performed as theater (silence = STOP)

Route to the right section of ${CLAUDE_PLUGIN_ROOT}/HARNESS.md:

  • Sequencing a task → §3 Steps. Plan-first is the highest-leverage habit.
  • Designing a loop/test → §5 Loops. It must go red when the thing is truly broken (test it broken first), be deterministic, fast, and agent-runnable. Red-first: write the failing loop before the code that makes it green.
  • Tools / permissions / injection defense → §4. The defense against prompt injection is the tool list, not a polite prompt — remove the dangerous tool.
  • Gates / commit boundary / approvals → §6. What must be shaped → prose; what must be guaranteed → a PreToolUse hook or deny-list. This repo ships a demonstrator commit-gate hook in hooks/.
  • Which model? → §7. Loops permit downgrade; missing loops force upgrade. Routing anxiety means a loop is missing. Cross-check PLAYBOOK.md §8 (cost per correct task; a tight spec beats a bigger model).
  • Session / handoff ritual → §8. Sync a beat early, before the window rots.
  • Something shipped wrong-but-green → §10 failure modes: config-check theater, green-test overconfidence, role confusion, assumption-as-fact, stale-context, dumb-zone drift, theater gate.

The loop quality test (apply before trusting any loop)

  • Does it go red when the thing is actually broken?
  • Could it return green while broken? (false-green trap)
  • Does it distinguish "correct" from "not found / permission denied"?
  • Can the agent run it without a human?

Read the relevant HARNESS.md section now, then design against it.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free