Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/studiokjm/ai-harness-template/learngit clone --depth 1 https://github.com/studioKjm/ai-harness-templateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/studiokjm/ai-harness-template/learn)<a href="https://agentmods.dev/commands/studiokjm/ai-harness-template/learn"><img src="https://agentmods.dev/badge/commands/studiokjm/ai-harness-template/learn.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00023 | $0.01343 |
| Opus 5 | $0.00012 | $0.00672 |
| Sonnet 5 | $0.00005 | $0.00269 |
| Haiku 4.5 | $0.00002 | $0.00134 |
Grade A, and why
learn scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/learn — Record and Promote Spike Findings
A spike without a learning record is just lost time. Make the knowledge durable.
When to use
- Right before closing a spike (
/spike close) — capture findings while context is fresh - After reviewing a learning to promote it to a permanent artifact (ADR, seed, code)
Usage
/learn record <spike-id> # Create learning.yaml from spike findings
/learn show <learning-id> # Print learning
/learn list [--spike <spike-id>] # List all learnings (optionally filtered)
/learn promote <learning-id> --to <adr|seed|code> --target <id>
# Mark learning as applied to a specific artifact
Prerequisites
record: spike must exist inspikingorlearnedstatepromote: learning must exist; target artifact must exist (ADR file, seed-vN, or code path)
Instructions
Step 1 — /learn record
For a recording invocation, the agent (you) does most of the work, not a script:
- Load the spike:
.harness/exploration/spikes/<spike-id>/spike.yaml - Read the sandbox:
.harness/exploration/spikes/<spike-id>/sandbox/— note files, measurements, code patterns - Generate a learning ID:
ln-<date>-<slug>(slug derived from spike question) - Bootstrap from template:
.harness/methodologies/exploration/templates/learning.yaml - Fill in:
id,created_at,spike_id,question(copy from spike)finding.summary— your 2-3 sentence answerfinding.confidence— high/medium/low (be honest)hypothesis_check— compare spike's hypothesis to actual finding (the most valuable field)evidence— at least one concrete artifact (file path, measurement, error message)recommendation.action— adopt | reject | defer | further_spikeapplies_to— where in the project this finding applies
- Save to
.harness/exploration/learnings/<learning-id>.yaml - Update spike's
links.learning_idto point at the new learning
Step 2 — Reporting after /learn record
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 131 lines · 23 tokens per session scan A 94cbd648ec40
learn is a command published in the GitHub repository studioKjm/ai-harness-template (43 stars, last pushed 3mo ago), licensed MIT. It adds 23 tokens to every session and 1,343 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
VibeGuard: ExecPlan
Long-term task execution plan — generates self-contained execution documents from SPEC, supports cross-session recovery.
VibeGuard: Review
Structured code review - first run the guard to obtain the baseline, then review according to security → logic → quality → performance priority.
VibeGuard: Live Truth
Verify live claims with fresh facts, inferences, and unresolved gaps.
capture-feedback
Quick feedback capture with structured signals.
check-gates
Run a Pre-Action Gate check against prevention rules before executing a risky action.
harness-adopt
Apply Harness Starter Kit to the current target repository with the prompt-first adoption workflow.