Harness Inbounds Checker — Development agent for Claude Code
Fresh-context invariant enforcer for the benchmarking climb loop.
How to install Harness Inbounds Checker
Installs to ~/.claude/agents/leadgrowgtm-loop-engineer-harness-inbounds-checker.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/LeadGrowGTM/loop-engineer/HEAD/.claude/agents/harness-inbounds-checker.md -o ~/.claude/agents/leadgrowgtm-loop-engineer-harness-inbounds-checker.md Restart Claude Code, or start a new session, for it to be picked up.
What Harness Inbounds Checker does
name: harness-inbounds-checker description: Fresh-context invariant enforcer for the benchmarking climb loop. Diffs a proposed variant against the declared invariant list BEFORE any measurement is spent. Did NOT invent the variant and has no view of the inventor's reasoning. Returns a binary IN-BOUNDS | VIOLATION verdict with the cited invariant. Use as the first pre-measurement check in every climb cycle. tools: Read, Glob model: claude-sonnet-5
You are the Harness In-Bounds Checker. Y
Alternatives in Development
- Eval Engineer — GAIA evaluation framework specialist 1.5k ★
- Peer Reviewer Demographics — A simulated peer reviewer specializing in population representativeness, demographic measurement, and populati 134 ★
- Corpus Prover — Proves a change on the maintainer's real local corpus — what number moved, by how much, and that nothing else 96 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Harness Novelty Checker
Fresh-context duplicate detector for the benchmarking climb loop. Diffs a proposed variant against the variant
Evidence Reconciler
Reconciles OncoKB vs CiVIC evidence for a single somatic variant and emits a unified AMP/ASCO/CAP tier call wi
Rules Enforcer
Owns the rulebook and the calendar. Deadlines, double and blank gameweeks, chip windows and legality, suspensi
DB Migration Writer
The ONLY agent allowed to write files. Used exclusively by the fix skill (the /claude-db:fix command) AFTER th
Gan Generator
Implement the plan produced by gan-planner against the named files. Returns proposed changes as unified diffs
Gemini Precheck
Pre-filter before the final Sol review. Has Gemini read a finished diff or subagent work to find obvious defec
Related Skills
Benchmarking Loop
Run a benchmark spec through the benchmarking loop (sweep or climb) and write a ranked variant ledger
Apply Diff
Apply a human-approved character diff from experience/journal/proposed-character-diffs/ to self/character.md.
AWS AI Analysis Agent
Serverless agent that reviews AWS Glue Jobs and Lambdas with Claude Haiku on Amazon Bedrock: findings with pro