skills/getcargohq__cargo-skills
1
0
Fork 0
mirror of https://github.com/getcargohq/cargo-skills.git synced 2026-09-14 19:17:15 +08:00
Code Issues Packages Projects Releases Wiki Activity
167 Commits 109 Branches 0 Tags
docs/script-require-allowlist
Commit Graph

4 Commits

Author SHA1 Message Date
Auree Aubert 69f8642cb5 cargo-analytics 1.5.0: describe what the run exports actually return (#98) 
* cargo-analytics 1.5.0: describe what the run exports actually return

`run download` was documented as returning "each run as a JSON object with
status, timing, executions, and runContext.<nodeSlug> containing per-node
outputs". It returns a signed URL to a gzipped CSV with one column per node
slug holding that execution's `title` — the truncated summary the diagnostics
runbooks call "never evidence". No runContext, no executions, not JSON.

- Rewrites the run download section around the real columns, with a warning
  to treat it as a status board across runs, not as node output.
- Corrects run download-outputs: --output-node-slug is optional (falls back to
  the last executed node), each row is input + output, and without --limit the
  file covers every matching run.
- Adds a section answering "can I get runContext for several runs at once" —
  you can't: it is a per-run S3 object and run get is the only command that
  hydrates it. Orchestration SQL is ruled out explicitly.
- Marks --is-finished broken in the six snippets that taught it; it returns
  400 unrecognized_keys on all four run commands. Replaced with
  --statuses success,error.
- Fixes the partial-retry pattern in polling.md, which piped run download
  stdout into jq '.recordId' — impossible against a signed URL to a gzipped
  CSV whose column is _record_id.
- response-shapes.md gains entries for both run exports and corrects batch
  download, which also returns {"url": …} rather than streaming to stdout.
- Adds the 500 case to analytics troubleshooting.

Verified against CLI 1.0.52 and the backend source. Underlying defects are
fixed in getcargohq/cargo#5579; these docs describe today's behavior.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* Drop the --is-finished caveats; the fix is a backend deploy

The caveats dated the docs against a deploy the bundle can't observe, and my
stated reason for them was wrong: cargo#5579 changes no CLI code for
isFinished (0 occurrences in its packages/cli diff) — the flag was always
sent, the API route schemas rejected it. So the fix lands when the backend
deploys, independent of the cli-version pin.

Examples keep --statuses success,error, which expresses the same filter and
is correct before and after that deploy, so nothing here expires. The specific
400 is replaced by a generic `400 unrecognized_keys` troubleshooting row that
stays useful whatever the flag, and points at report create.

Also drops "until the fix ships" from the 500 row, for the same reason.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* Restore --is-finished: it is wider than success+error

Swapping the flag for `--statuses success,error` was semantically wrong, not
just verbose. `isFinished` maps to `finished_at IS NOT NULL`, and cancelled
(cancelRuns.ts:94) and skipped (launchRuns.ts:151) runs stamp finishedAt too.
Skipped is routine — checkToSkipRuns marks records skipped as ordinary dedupe
behaviour — so the substitution would silently under-report on any play that
skips records.

exports.md and run-analytics.md revert to their original content: their
--is-finished examples were correct all along. The run download section now
shows both forms with a note that they are not interchangeable.

The flag still 400s until cargo#5579 reaches production, which is a backend
deploy independent of the CLI release; the generic 400 unrecognized_keys row
in troubleshooting.md covers that window without dating the docs.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-08-14 17:24:10 -07:00
Auree Aubert 3708f8c955 Agent discoverability: trigger-shaped descriptions, routing evals, standalone bootstrap (#77) 
* Agent discoverability: trigger-shaped descriptions, routing evals, standalone bootstrap

The `description` field is the only text an agent weighs before deciding
whether to load a skill. Everything else in the bundle is invisible until
that decision is already made — and twelve of the seventeen descriptions
read as "Manage <CLI nouns> using the Cargo CLI", which matches a user who
is already thinking in Cargo's object model rather than one stating a job.

This rebuilds that layer and, for the first time, tests it.

Routing
- Every description rewritten to a four-part template: job in the user's
  words -> literal quoted trigger phrases -> proper nouns -> an explicit
  `Skip when:` pointing at the sibling skill that owns the adjacent case.
- Integration and provider names are generated, not hand-maintained.
  sync-trigger-slugs.ts fills cargo-gtm's `Providers:` list from the
  playbook tree and cargo-connection's `Integrations:` list from a
  committed catalog snapshot (138 integrations, 50 credits-based).
  `--refresh` re-pulls from a logged-in CLI; `--check` gates CI.
- evals/routing.jsonl (92 prompts) + routing-eval.ts, wired into CI.
  Structural tier asserts the template and that each skill keeps >=5 terms
  unique to it; lexical tier requires the expected skill to rank first;
  an opt-in --llm tier asks a real model to route from descriptions alone.
  Deep paraphrases are tagged "hard" and reported rather than gated — a
  lexical proxy cannot judge them fairly.

Standalone installs
- `skills add --skill <one>` is a supported path and the registry lists
  each skill separately, but thirteen skills' only install instructions
  were a relative link to ../cargo/references/prerequisites.md, which does
  not exist in a single-skill install. Every SKILL.md now has a
  self-contained Bootstrap section, enforced by the linter.

Coverage
- New capability skill cargo-segmentation — the segmentation CLI domain,
  previously uncovered while segments were load-bearing in most recipes.
  Verified against the live CLI, including the flag traps.
- Three cargo-gtm recipes: source-planning (probe and cost candidate
  sources per *hit* before any fan-out), ads-audience-activation (Google
  Customer Match / LinkedIn Matched Audiences), review-and-iterate (human
  review of judgment output, kept as an eval set).
- Router gains a "these skills vs a workspace MCP server" section: Cargo
  has no first-party MCP server — a workspace builds one and serves it
  over `cargo-ai mcp` — so the routing rule is scale and reproducibility.

Index and hygiene
- llms.txt now lists every recipe and provider playbook with a generated
  summary: 17 entries -> ~80 job-named addressable URLs.
- Router trimmed ~80 lines of per-skill summaries that restated each
  skill's own description; critical rules kept in full.
- Linter: bootstrap check, skill-count claims validated against the tree,
  `doctor` and `mcp` added to the known CLI domains.
- CONTRIBUTING.md, keyword-dense README opening, updated repo metadata.

Also carries the aiArk/apolloio priority-stack promotion authored in a
parallel session on this branch (cargo-gtm 1.12.0) — the two touch the
same SKILL.md files and could not be split cleanly.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* Surface the 100-credit free tier where an agent meets it

There is no purchase gate between installing Cargo and producing a real
deliverable — a new account starts with 100 free credits and no card — and
nothing in the bundle said so. An agent that doesn't know the budget can't
tell a first-time user whether the demo is a look around or a commitment,
and defaults to hedging.

Now stated at every point an agent encounters the question:

- README opening and the generated llms.txt install block: an agent can
  install, sign the user up, and hand back a list in one turn.
- Router install section and prerequisites.md, with anchors for what 100
  credits buys (~5,000 sourced leads, ~1,000 verified-email enrichments,
  ~50 fully enriched contacts).
- cargo-quickstart: the demo spends ~0.5 of the 100, and the skill now
  says so before the first paid call — that reframes the moment from a
  purchase decision into a look around, which is the job of a quickstart.
- cargo-billing: a "what 100 credits buys" table, plus two new trigger
  phrases so free-tier questions route here at all.
- cost-discipline.md: on a new account, receipts and next-step proposals
  must be framed against the free tier. "87.6 of your 100 free credits
  left" is a number the user can act on; "87.6 remaining" is not. Also
  warns that phone lookups (6-7/call) are the fastest way to burn a tier
  a first session should end with something left to try.

Two routing eval cases added for free-tier questions (79/79 core).

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* Run the routing evals against a real model in CI

The lexical tier passes 79/79, but it is a deterministic proxy written
alongside the descriptions it grades — it cannot fail in a way that
surprises us, which is what makes it weak evidence. The only honest
scoreboard for the description rewrite is whether a real model, shown
nothing but the 17 descriptions, routes the eval prompts correctly.

That tier existed but had never run. Now it does, in its own workflow.

- New .github/workflows/routing-evals-llm.yml, using the repo's
  ANTHROPIC_API_KEY. Deliberately not part of skills-lint: it costs money
  and is non-deterministic, so it must never stand between a typo fix and
  a merge. Triggers on PRs and pushes that touch a SKILL.md, the eval
  cases, or the runner; weekly on a schedule so model drift surfaces even
  when nobody edits a description; and on demand with a model override.
- runLlm reworked for CI: 6-way concurrency (92 sequential calls is a slow
  job), retry with backoff on 429/5xx, tolerant answer parsing, and a
  core/hard/overall breakdown instead of one number.
- Results are written to GITHUB_STEP_SUMMARY as a table, including every
  misroute with what the model picked instead — so a regression is
  readable on the run page without opening logs.
- Optional --llm-min=<percent> floor, wired to the exit code. Left unset
  until the first runs establish a baseline; a floor guessed before the
  data would either never fire or block on noise.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* Act on the first real-model routing run: 92/94, two fixes

Baseline from the model tier (claude-haiku-4-5, judging from the 17
descriptions alone): 92/94 overall — core 79/79, hard 13/15. The lexical
proxy's 3/15 on the hard tier was noise about the proxy, not the
descriptions, which is what that tier existed to establish.

Both misroutes were worth having:

- "Our outbound is going to the wrong people — help me figure out who
  actually buys from us" routed to cargo-context instead of cargo-gtm.
  A real collision: cargo-context's triggers include "who do we sell to",
  which reads as ICP ownership. It owns *writing the conclusion down*;
  deriving it from won/lost data is cargo-gtm. Added that boundary to
  cargo-context's Skip when, and the two phrasings to cargo-gtm.

- "Give the assistant access to our sales collateral so it stops making
  things up" routed to cargo-content instead of cargo-ai — and the model
  was right. As written the first step is getting the collateral into the
  workspace. Corrected the expectation and added the other side of the
  boundary as its own case ("attach the library we already uploaded to
  the agent"), which the lexical tier then failed, correctly: cargo-ai
  never said "library" or "release resources". Added both, so the
  content/ai split is now legible to a ranker and a model.

Set --llm-min=93 now that a baseline exists. Deliberately below the
observed 98%, not at it: a model tier that fails on one non-deterministic
flip trains people to ignore it.

Lexical 80/80 core.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* Document cargo-ai mcp — the bridge that was written down nowhere

Cargo reaches coding agents through two doors. Skills reach agents that
implement skills.sh; a workspace MCP server reaches everything else —
Claude Desktop, ChatGPT, and whatever ships next. The second door was
fully built and documented in exactly one place: `cargo-ai mcp --help`.

Worse, cargo-ai/SKILL.md taught `ai mcp-server create` (building a server)
without ever mentioning the command that serves it, so an agent could
construct one and have no idea how to connect it — and the section
conflated publishing a server with consuming someone else's, which are
unrelated surfaces.

- MCP section rewritten around the two directions, with a table: publish
  (`ai mcp-server`, called by any MCP client, wired via `cargo-ai mcp`)
  vs consume (`ai mcp-client`, called by your agents, wired via
  `release update-draft --mcp-clients`). States plainly that there is no
  first-party Cargo MCP server — a workspace builds its own.
- Documents `cargo-ai mcp` with the registration one-liner, the
  CARGO_MCP_SERVER_UUID fallback, and the stdout-is-protocol constraint.
- Details verified against the live API rather than --help: actions take
  kind "tool" *or* "agent" (an agent is exposable as a callable MCP
  tool), resources take model or file, and `update` replaces the arrays
  wholesale instead of merging — a real footgun.
- Routing rule for which door to use, pointing at the router section.
- Two trigger phrases ("expose our tools as an MCP server", "use Cargo
  from Claude Desktop or ChatGPT") and two eval cases, so the publish
  direction is reachable from a description at all. It previously wasn't.

Lexical 81/81 core.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
 2026-08-12 10:18:32 -07:00
replicas-connector[bot] e28aabd521 Point the skills at browserless login, and unpin agents from a CLI that lacks it (#73) 
* Point the skills at browserless login, and unpin agents from a CLI that lacks it

cargo-ai 1.0.47 (2026-08-10) added `login --email`: it mails a one-time code,
creates the account on first use, and needs no browser at any point. All 16
skills still described sign-in as "browser sign-in via --oauth, or an API
token".

The load-bearing change is `cargo/cli-version`. The router tells every agent to
run `npm install -g @cargo-ai/cli@$(cat cli-version)` at session start, and that
pin was 1.0.29 (2026-07-09). Its `login` command exposes only --oauth, --token
and --workspace-uuid, so documenting --email without moving the pin would ship a
flag the installed binary does not have.

- cargo/cli-version: 1.0.29 -> 1.0.47
- 16 compatibility lines rewritten
- router + README install blocks lead with --email, and name the agent-shell
  behaviour (no terminal: first call sends the code and exits, re-run with
  --code) and the stdin form
- patch version bump on each changed SKILL.md so clawhub republishes them

Co-Authored-By: Auree Aubert <16732193+aureeaubert@users.noreply.github.com>
Co-Authored-By: replicas-connector[bot] <replicas-connector[bot]@users.noreply.github.com>

* Regenerate skill-metadata.json for the changed skills

The bundle linter tracks a version and a contentHash per skill and failed with
19 stale entries. Regenerated with the documented command, so the diff is only
the bumped versions and their matching hashes.

Co-Authored-By: Auree Aubert <16732193+aureeaubert@users.noreply.github.com>
Co-Authored-By: replicas-connector[bot] <replicas-connector[bot]@users.noreply.github.com>

* Mirror the router version bump into the three plugin manifests

The linter requires .claude-plugin, .codex-plugin and .cursor-plugin to carry
the router's version. 1.17.0 -> 1.17.1.

Co-Authored-By: Auree Aubert <16732193+aureeaubert@users.noreply.github.com>
Co-Authored-By: replicas-connector[bot] <replicas-connector[bot]@users.noreply.github.com>

---------

Co-authored-by: replicas-connector[bot] <replicas-connector[bot]@users.noreply.github.com>
Co-authored-by: Auree Aubert <16732193+aureeaubert@users.noreply.github.com>
 2026-08-11 14:42:02 -07:00
Auree Aubert 286aaaf981 feat: prompt library + generated skill metadata (#55) 
Rebased onto main (post #53/#54 squash): 38 named parameterized LLM prompts in
six sharded files with a grep-first index (temperature guidance scoped to the
bulk tier; judgment-tier models that reject non-default sampling get no
override), plus a generated skill-metadata.json per skill — typed doc catalog
+ deterministic content hash (code-unit ordering, locale-independent),
produced by .github/scripts/skills-metadata.mjs (--write) and verified by
skills-lint.mjs so stale metadata fails CI. Metadata regenerated over main's
tree, which is why lint was failing before: the catalogs predated main's
email_status recipe fix. cargo-gtm -> 1.4.0.

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
 2026-07-10 00:26:04 -07:00