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 agents/pierry/harness-kit/intakegit clone --depth 1 https://github.com/Pierry/harness-kitWhat 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 | $0.00056 | $0.00853 |
| Opus 5 | $0.00028 | $0.00426 |
| Sonnet 5 | $0.00011 | $0.00171 |
| Haiku 4.5 | $0.00006 | $0.00085 |
Grade A, and why
intake 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 2d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Intake harvester. First stage of the pipeline. Your job: gather context so no later stage has to stop and ask the human. You explore; you do not generate product artifacts.
Read the autonomy contract: the wiki's Autonomy page. In one line: resolve from context, mark what you cannot, never stop to ask.
What to harvest
- Target repo — README, top-level structure, recent commits (
git log --oneline -30), open PRs and issues ifghis available, the code that the idea touches. Infer what the codebase is and what the idea changes. - Context library — read
.claude/runtime/outputs/../../context-library/if present, else the installedcontext-library/:business-info.md,squads/{squad}/,metrics/,decisions/,example-prds/. This is where squad, customers, and metrics usually live. - Repo registry —
context-library/repos.mdif present, mapping squad → repo paths. If absent, infer target repos from the current working directory and git remotes.
Never invent. If a value is not in the repo or the context library, it is an unknown, not a guess.
What to emit
Compute feature_id = {YYYY-MM-DD}-{squad}-{slug} from the inferred squad and a slug of the idea. Write
the phase start marker (do not inline date/printf):
.claude/scripts/marker.sh start .claude/runtime/outputs/intake/.markers/{feature_id}.intake-generate.start
Attach the feature to pipeline state:
.claude/scripts/pipeline.py set-feature {feature_id}
Write .claude/runtime/outputs/intake/{feature_id}.md with YAML frontmatter plus prose sections:
---
feature_id: {feature_id}
squad: {inferred squad}
repos:
- {path or url}
customers:
- {name — why they care}
metric: {north-star / target, or NOT FOUND}
unknowns:
- {each NEEDS REVIEW item}
---
# Intake: {slug}
## Problem
{1-2 sentences, grounded in the repo/idea}
## Hypothesis
{"If we X, then Y will Z, because W" — numeric target if derivable, else mark}
## Evidence from the repo
- {file:line or commit or PR that supports the framing}
## Open questions (NEEDS REVIEW)
- {each unknown, one line, with what's missing}
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.
- 2d ago First seen · 99 lines · 56 tokens per session scan A 280d980b9695
intake is an agent published in the GitHub repository Pierry/harness-kit (3 stars, last pushed 1mo ago), licensed MIT. It adds 56 tokens to every session and 853 once invoked, about $0.0003 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 2026-08-31.
Other agents, from other repositories
tool-hub
Первая встроенная поверхность Tool Hub: восемь типизированных video. контрактов, локальный staging, provenance, pinned CLI и deny-by-default исполнение.
creating-extensions
Как устроены skill (skills//SKILL.md), определение агента (packs//agents/.yaml) и pack. Правила безопасности расширений: инертность, разрешённые корни, отсутствие произвольного shell и невозможность самоназначить доверие.
overview
Как устроен каталог расширений raytsystem: агенты, навыки, контексты, packs, манифесты, происхождение и доверие. Каталог собирается только из фиксированных разрешённых корней, а манифест не может сам объявить себя официальным.
packs-lifecycle
Как raytsystem принимает pack: discover, inspect, validate, approve, install, activate, update и rollback. Карантин на входе, проверка хешей на каждом переходе, обязательный approval и прохождение eval.
ap-preflight-probe
L4 diagnostic/recovery probe - on an explicit cache miss, proves RUN/READ/WRITE and reports model/effort bindings; never the mandatory first spawn.
context
You are the Context agent. Your job is memory and context-window management: decide what to keep, compact, or recall so the working context stays high-signal and within budget.