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 skills add sabbour/agentweaver --skill api-harnessgit clone --depth 1 https://github.com/sabbour/agentweaverWrote 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/skills/sabbour/agentweaver/api-harness)<a href="https://agentmods.dev/skills/sabbour/agentweaver/api-harness"><img src="https://agentmods.dev/badge/skills/sabbour/agentweaver/api-harness.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 5.1 | $0.00000 | $0.01776 |
| Opus 5 | $0.00000 | $0.00888 |
| Sonnet 5 | $0.00000 | $0.00355 |
| Haiku 4.5 | $0.00000 | $0.00178 |
Grade A, and why
api-harness 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 — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
API harness CLI contract
Use the API harness to drive Agentweaver through its REST API, capture the complete
request/response evidence, and emit a normalized
agentweaver.persona-judge-verdict/v1 JSON verdict. It is for backend/API
end-to-end validation; use the UI or MCP harness for those surfaces.
Run all commands below from the repository root. The harness requires Node 18 or
newer. It requires an access token from the transient AGENTWEAVER_TOKEN
environment or the explicit recorder-session provider. It never accepts bearer
material in process arguments and never borrows gh auth token or
GITHUB_TOKEN for a remote target.
Token acquisition for staging (Entra Conditional Access)
Agentweaver staging uses Entra Conditional Access. The token cannot be retrieved via device-code flow or plain browser. With an already authenticated demo-recording session, use the recorder-session provider. It reads the protected session token only in memory for each request; never export it to an environment variable, CLI argument, transcript, finding, verdict, or log:
node scripts/api-harness/run-persona.mjs `
--scenario generated-artifacts-seam `
--target https://<host>.staging.<domain> `
--auth-provider recorder-session
The provider uses scripts/demo-recording/.auth/ by default; use
--recorder-auth-root only for an existing protected recording-auth root. Verify the
recording session first with npm run demo:record -- status. If its bearer is expired,
the human-only recording sign-in flow must refresh it.
Driving a persona scenario (the only way — dynamic, no fixed scripts, no HTTP-calling wrapper)
There is no curated list of named scenario subcommands, no per-persona fixed
step sequence, and no scripted HTTP-calling layer standing between the driving
actor and the target. Harness dispatches a fresh PersonaActor sub-agent
(.github/agents/persona-actor.agent.md — a real CLI agent with shell/execute
access) to fully impersonate the persona in an isolated context — handed ONLY
that persona's brief — scripts/persona-briefs/personas/<id>.md +
scripts/persona-briefs/surfaces/<id>.api.md — and it decides every next action
live by calling whatever operation it resolves from the real API's OpenAPI/
Swagger document, one turn at a time, reacting only to the real response it gets
back. This is what lets the actor push back, poll events, and adapt — a fixed
script (or a fixed subcommand wrapper) structurally cannot.
What ships with it
24 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- findings/.gitignore 69 B
- JUDGE.api.md 391 B
- lib/auth-providers/recorder-session.mjs 812 B runs code
- lib/client.mjs 5.5 KB runs code
- lib/generation-checks.mjs 10 KB runs code
- lib/metrics.mjs 3.0 KB runs code
- lib/reporter.mjs 6.0 KB runs code
- lib/seams.mjs 13 KB runs code
- package-lock.json 771 B
- package.json 345 B
- README.md 5.3 KB
- run-persona.mjs 16 KB runs code
- scenarios/generated-artifacts-seam.mjs 2.8 KB runs code
- test/auth-providers.test.mjs 1.6 KB runs code
- test/generation-checks.test.mjs 7.2 KB runs code
- test/run-persona.test.mjs 2.1 KB runs code
- test/seams.test.mjs 3.0 KB runs code
- transcripts/.gitignore 85 B
- transcripts/memory-agent-roundtrip-20260715T014105.judge-prompt.txt 7.5 KB
- transcripts/memory-write-verify-20260715T010957-judge-prompt.txt 12 KB
- transcripts/oracle-live-20260714T191315.judge-prompt.txt 3048 KB
- transcripts/skill-usage-verify-20260715T015920-judge-prompt.txt 9.8 KB
- transcripts/wave1-verify-20260715T004112.judge-prompt.txt 16 KB
- verdicts/.gitignore 77 B
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 Changed · +4 lines c9deab468f28
- 4d ago First seen · 150 lines · 0 tokens per session scan A ef3678da717d
api-harness is a skill published in the GitHub repository sabbour/agentweaver (5 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,776 tokens. 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 2026-09-03.
Other skills, from other repositories
verify
Build, launch, and drive the AppointMe stack to verify a change end-to-end at its runtime surface.
api-test
Run API integration tests against the running backend, verify endpoints return expected responses and status codes. Use after deploying a preview or starting the dev server.
api-test
Run API integration tests against the running backend, verify endpoints return expected responses and status codes. Use after deploying a preview or starting the dev server.
webapp-testing
Start/reuse a local app, wait for readiness, inspect rendered state/console/network, act from observed selectors, and verify with evidence.
signal-end-to-end
Implements end-to-end signal changes spanning router config, signal extraction, CLI schema, optional bindings, router-owned metadata headers, and E2E test coverage. Use when adding a new signal type, changing signal configuration or extraction logic, updating CLI schema for signal parameters, or modifying router-owned…
qa
Browser-based QA verification. Launches a real browser, navigates the app, clicks buttons, fills forms, and tests user flows. Works as a standalone skill or as a phase end condition in campaigns. Requires Playwright (optional dependency, graceful skip if not installed).