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/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/sabbour/agentweaver/api-harness"><img src="https://agentmods.dev/badge/skills/sabbour/agentweaver/api-harness.svg" alt="Reviewed on agentmods" width="80" 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 | $0.05163 |
| Opus 5.5 | $0.00000 | $0.02065 |
| Sonnet 5.5 | $0.00000 | $0.01033 |
| Haiku 4.5 | $0.00000 | $0.00516 |
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 19d 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 — 400 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. At its first authenticated call, it obtains the complete Authorization value only through the cached
UI-harness recorder-session provider, in memory, and never accepts bearer material in
process arguments or borrows
gh auth token or GITHUB_TOKEN for a remote target.
Browser-managed authentication for staging (Entra Conditional Access)
Agentweaver staging uses Entra Conditional Access. Before an authenticated API run,
run the documented demo-recording session setup once for that target. The
recorder-session provider reads its cached UI storage/session sidecar, verifies it
belongs to the requested origin, and returns the complete Authorization value only in
memory. Pass that value to the header unchanged; do not add another scheme. It never
starts a second browser sign-in or exports the value 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>
The provider uses scripts/demo-recording/.auth/recording.storageState.json and
its .sessionStorage.json sidecar by default; use --recorder-auth-root only
for an explicitly chosen protected auth root. To reuse a supported UI login:
node scripts/api-harness/run-persona.mjs --scenario generated-artifacts-seam --target https://<host>.staging.<domain> --recorder-auth-root scripts/ui-harness/.auth
run-oracle-acceptance.mjs accepts the same --recorder-auth-root; the context
budget CLI uses --auth-root scripts/ui-harness/.auth. In an explicit root,
recording.storageState.json takes precedence if both layouts exist; otherwise
staging.storageState.json and its matching seed are used. No search in another
directory or fallback after an invalid selected cache occurs.
--auth-provider recorder-session remains accepted for clarity but is the default.
If the cached recorder session is absent, expired, or belongs to another origin, it fails
with npm run demo:record -- open --base-url <origin> for the recorder layout,
or node scripts/ui-harness/login-chrome-default.mjs --base-url <origin> for the
selected UI layout. Microsoft
Entra account selection, credentials, MFA, and consent remain human-only.
For a decoded JWT with an exp claim, a provider reuses the in-memory
Authorization value only while it remains unexpired. After expiry it revalidates
the same selected cache path and origin; a refreshed session can be reused without
creating a new provider, but an expired or wrong-origin replacement fails before
an API call. Tokens without a decodable expiry retain the existing behavior.
Do not fall back to generic Playwright, direct CDP/DevTools, ad-hoc profile
launch/copy, or manual browser automation; surface the provider's recovery error and
use the demo-recording login/cached-session flow. That flow requires the installed literal
Google Chrome chrome.exe on Playwright's chrome channel, never bundled Chromium.
What ships with it
36 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
- issue-1603-live-retest.mjs 17 KB runs code
- JUDGE.api.md 391 B
- lib/auth-providers/local-test.mjs 742 B runs code
- lib/auth-providers/recorder-session.mjs 4.0 KB runs code
- lib/client.mjs 6.2 KB runs code
- lib/context-budget-pressure.mjs 7.6 KB runs code
- lib/deployment-sha.mjs 1.6 KB runs code
- lib/generation-checks.mjs 22 KB runs code
- lib/metrics.mjs 3.0 KB runs code
- lib/oracle-acceptance.mjs 41 KB runs code
- lib/reporter.mjs 5.9 KB runs code
- lib/seams.mjs 44 KB runs code
- package-lock.json 771 B
- package.json 345 B
- README.md 13 KB
- run-context-budget-pressure.mjs 3.1 KB runs code
- run-oracle-acceptance.mjs 4.7 KB runs code
- run-persona.mjs 16 KB runs code
- scenarios/generated-artifacts-seam.mjs 4.7 KB runs code
- test/auth-providers.test.mjs 15 KB runs code
- test/context-budget-pressure.test.mjs 5.3 KB runs code
- test/deployment-sha.test.mjs 2.8 KB runs code
- test/generation-checks.test.mjs 13 KB runs code
- test/local-auth-provider.test.mjs 1.8 KB runs code
- test/oracle-acceptance.test.mjs 56 KB runs code
- test/oracle-openapi-contract.test.mjs 572 B runs code
- test/run-persona.test.mjs 4.9 KB runs code
- test/seams.test.mjs 31 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.
- 19d ago Changed · +12 lines a734c7a290d9
- 21d ago Changed · +25 lines 03b54e481cd2
- 23d ago Changed · +73 lines b5e5e72c9d7d
- 28d ago Changed · +2 lines fad9c9ced7b0
- 29d ago Changed · +22 lines eb3e904b9297
- 1mo ago Changed · +24 lines c902d2d586b1
- 1mo ago Changed · +34 lines fe9d065b8c11
- 1mo ago Changed · +24 lines fdce26752146
- 1mo ago Changed cbd7ecfd0448
- 1mo ago First seen · 184 lines · 0 tokens per session scan A a7ac619cc499
api-harness is a skill published in the GitHub repository sabbour/agentweaver (7 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 5,163 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-testing
Generate and run Postman collections, validate API responses, manage test suites.
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.
mirrord-chaos
Help users chaos test their app with mirrord: inject artificial latency or connection errors into a mirrord session's outgoing traffic via per-session chaos rules managed with the mirrord chaos CLI. Use when a user wants to add latency or delay to outgoing connections or a dependency (e.g. a slow database), simulate…