Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Harness Improvement · kdlbs | DecimalAI

agent toolingby kdlbs

Harness Improvement

kdlbs/harness-improvement

Source kdlbs/kandev ↗License AGPL-3.0

Improve Kandev's AI harness from session learnings or explicit requests.

SkillScore89Maintained

Automatic

AI Engineer pack

Last verified 10/9/2026

What it changes

Passed 16 of 18 test cases with the skill, and 8 of 18 without it.

Without skill44%

With skill89%

Checks it passed that the model alone missed

Uses apply_patch for making file edits.
Avoids executing a generic harness-improvement workflow.
Requires adapting tool names and commands to Kandev conventions.

and 7 more in the full report

8 fixed8 passed either way2 still failing0 broke4 errored

Pass-rate lift: +44 pp
Tokens: +24%
Agent turns: 0%
Response time: -45%

Model gemini-3.6-flashMeasured 10/9/2026v416 of 22 cases passed with skill · 18 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/kdlbs-harness-improvement/SKILL.md https://app.decimal.ai/s/kdlbs-harness-improvement/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add kdlbs-harness-improvement, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/kdlbs-harness-improvement/ and pins it in .decimal/skills.lock

decimalai skills add kdlbs-harness-improvement

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/kdlbs-harness-improvement/SKILL.md https://app.decimal.ai/s/kdlbs-harness-improvement/SKILL.md

Raw file: /s/kdlbs-harness-improvement/SKILL.md ↗ · machine-readable summary at /s/kdlbs-harness-improvement.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 8

Prompt

We renamed `old-skill` to `new-skill` across our repository. What search tool command should we run to verify that no stale references to `old-skill` or `old-agent` remain in AGENTS.md or .agents?

What was checked, without and with the skill
Checked	Without skill	With skill
Executes ripgrep to search for stale identifier names across instruction files.	Missed	Passed

case-13 · gemini-3.6-flash · 10/9/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md86 lines6 files

Harness Improvement

Use this skill to turn lessons from real agent sessions into durable harness changes: skills, agents, subagents, commands, scripts, and always-on instruction files.

For an explicitly requested session retrospective, use /retro to filter lessons and propose shared-file edits.

Planner Entry

The planner may inventory, edit, and validate a small localized harness change directly. Delegate broad cross-platform migration or independent work when it has positive ROI. Do not use Kandev MCP task/session APIs for workers.

Choose The Artifact

Before editing, classify the requested improvement:

Session learning: recurring failure, workaround, or convention discovered during a session. Read references/session-learnings.md.
Skill: task-specific playbook loaded on demand. Read references/skills.md.
Custom agent: an exception that requires the user to explicitly reverse the repository's single-session policy. Read references/agents.md before acting.
Command: explicitly invoked workflow shortcut. Prefer a skill unless the user wants manual invocation only.
AGENTS.md / CLAUDE.md / rules: always-on or path-scoped instruction. Read references/instructions.md.
Cross-platform migration: preserving behavior across Claude, Codex, Cursor, and OpenCode. Read the relevant platform files in references/platforms/.

If the user names a target platform for a skill, command, or instruction file, load only that platform reference. Do not create a custom agent or platform mirror unless the user explicitly requests a policy reversal.

Workflow

Inventory first
- Use rg --files to find existing .agents/skills, .claude, .codex, .cursor, .opencode, AGENTS.md, and CLAUDE.md files.
- For platform-specific formats, read the bundled files under references/platforms/ before consulting external docs. Treat those files as the first source of truth for Claude, Codex, Cursor, and OpenCode harness layout.
- Check for duplicate or superseded skills/agents before adding new ones.
- Prefer updating the existing artifact when the behavior belongs to an existing workflow.
Normalize the learning
- Convert anecdotes into reusable guidance: trigger, problem, correct action, fallback, verification.

Track record

Benchmark

16 / 18

test cases passed, vs 8 / 18 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 10/9/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

kdlbs / kandev

.agents/skills/harness-improvement/SKILL.md

View on GitHub ↗

License AGPL-3.0Imported 10/9/2026Version v4Bundled 5 files

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Harness Improvement

Planner Entry

Choose The Artifact

Workflow

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills