list-adapters
Skill298 starsChanged 6 months agoMITView on GitHub
Show every capability the engine knows about and which provider backs it (built-in TypeScript, bundled YAML, or user-installed YAML), with availability flags reflecting whether each provider's API key is set. Use when the user says 'list adapters', 'show me which providers are configured', 'which providers are available', 'what capabilities can YALC use right now', or 'show capability coverage'. Read-only — never writes anything.
In this repository37 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- answer-linkedin-commentsSkill
- build-routineSkill
- campaign-dashboardSkill
- campaign-strategySkill
- claap-weekly-recapSkill
- cross-platform-schedulerSkill
- debuggerSkill
- enrich-with-signalsSkill
- find-lookalikesSkill
- fullenrich-content-engagersSkill
- fullenrich-event-attendeesSkill
- fullenrich-network-activationSkill
- fullenrich-plg-reverse-lookupSkill
- icp-importSkill
- import-leadsSkill
- integrate-with-yalcSkill
- launch-linkedin-campaignSkill
- lemlist-campaign-from-icpSkill
- lost-deal-revival-agentSkill
- notion-reviewSkill
- personalize-messageSkill
- predictleads-dashboardSkill
- predictleads-lookalikesSkill
- predictleads-signalsSkill
- prospect-discovery-pipelineSkill
- provider-builderSkill
- qualify-leadsSkill
- research-prospectSkill
- run-competitive-intelSkill
- run-doctorSkill
- scrape-post-engagersSkill
- send-cold-emailSkill
- setupSkill
- show-routineSkill
- track-campaignsSkill
- yalc-orientationSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Use this skill when the user says:
- Distinctive: line 44
Onboarding interruption guard: Run:
- Very specific: line 95
If the JSON includes any declarativeErrors entries, render them after the inventory.
- Distinctive: line 99
If any capability has zero available providers, ask:
- Very specific: line 103
If a bundled provider exists but its key isn't set, suggest the relevant /keys/connect/<provider> URL via the dashboard command:
- Distinctive: line 107
Don't run anything unless they say yes.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
List Adapters
I'll show you every capability the engine recognizes and which provider currently backs it. This is the read-only inventory that proves YALC's adapter registry is correctly wired and tells you which capabilities are actually executable on your machine.
When This Skill Applies
Use this skill when the user says:
- "list adapters"
- "show me which providers are configured"
- "which providers are available"
- "what capabilities can YALC use right now"
- "show capability coverage"
NOT this skill (use provider-builder instead):
- "add a new provider for X" — provider-builder authors a fresh YAML manifest.
- "wire up [vendor] to YALC" — same.
NOT this skill (use run-doctor instead):
- "is YALC working" / "diagnose YALC" — run-doctor is the broader 5-layer health check; this skill only inventories adapters.
What This Skill Does
- Runs
yalc-gtm adapters:list --jsonto get the structured adapter registry. - Parses the JSON and renders it grouped by capability.
- For each capability, lists every provider the engine knows about with its source tag —
[built-in](TypeScript),[bundled](YAML shipped with YALC underconfigs/adapters/), or[user](YAML the user dropped in~/.gtm-os/adapters/). - Marks each provider with availability —
✓if every required env var is set,✗if any required key is missing. - Surfaces which provider wins resolution for each capability based on the user's
~/.gtm-os/config.yamlpriority list (or the default priority if no override).
What This Skill Does NOT
- Modify the registry, env files, or any config. Read-only.
- Smoke-test live vendor endpoints. That's
adapters:smoke <path>. - Auto-install missing providers. Use
provider-builder(author a new YAML) orprovider:install <capability>/<provider>(fetch fromproviders/manifests/).
Pre-flight (do this before step 1)
Onboarding interruption guard. Run:
test -f ~/.gtm-os/.in-flight-setup && echo "BLOCKED" || echo "OK"If BLOCKED, surface a soft warning: setup is mid-flight, so the adapter inventory may not yet reflect freshly-set keys or newly-installed manifests. Since this skill is read-only and never writes anything, ask the user whether to (a) run anyway against the in-flight state, or (b) finish yalc-gtm setup --resume first and re-invoke. Default to (b) unless the user explicitly opts in to (a).
Workflow
Step 0: No user input needed
This is a read-only command. Skip straight to step 1.
Step 1: Run the CLI
From the gtm-os root (~/Desktop/gtm-os/):
npx tsx src/cli/index.ts adapters:list --jsonThe --json flag emits a structured payload instead of the human-readable table. Per the 0.13.0 benchmark, single-command skills like this one shell out unconditionally — the import-direct path saves only ~10ms here, not worth the maintenance cost. Use the import-direct pattern only for chained skills (see docs/skills-architecture.md).
Step 2: Parse the JSON output
The CLI emits { rows: Array<{ capability, providers: Array<{ id, source, available, priorityIndex?, manifestPath? }> }>, declarativeErrors: Array<...> }.
capabilityis the capability id (e.g.,icp-company-search,crm-contact-upsert).providers[].sourceis'built-in','bundled', or'user'.providers[].availableistrueif every env var the adapter needs is set,falseotherwise.providers[].priorityIndexis the resolution rank (1 = first chosen). Providers without an index are registered but not in the priority list.declarativeErrorslists any YAML manifests that failed to compile at boot — surface these prominently if non-empty.
Step 3: Render a clean inventory
Group by capability. For each capability, list every provider in priority order followed by any non-prioritized entries. Use the format shown in references/example-output.md:
icp-company-search
#1 ✓ crustdata [built-in]
#2 ✓ apollo [built-in]
· ✗ pappers [built-in]#N is the priority rank. · marks providers registered but not in the priority list. ✓ / ✗ reflects availability.
End the summary with a one-line verdict: how many capabilities have at least one available provider, and how many don't (these are the gaps the user might want to fill).
Step 4: Surface declarative errors
If the JSON includes any declarativeErrors entries, render them after the inventory. Each error has the manifest path and the validation message. These are user-actionable — the user dropped a malformed YAML in ~/.gtm-os/adapters/ and the engine refused to compile it.
Step 5: Offer follow-ups
If any capability has zero available providers, ask:
"Want me to add a provider for
<capability>? I can runprovider-builderto author a new YAML manifest from a vendor docs URL."
If a bundled provider exists but its key isn't set, suggest the relevant /keys/connect/<provider> URL via the dashboard command:
"
<provider>is bundled but<ENV_VAR>isn't set. Want me to open/keys/connect/<provider>in your browser? I'll runyalc-gtm dashboard --route /keys/connect/<provider>."
Don't run anything unless they say yes.
Notes
- The CLI's
--jsonflag was added in 0.11.0 alongside the declarative adapter loader (B2). It's stable and parseable. - This skill never imports the registry directly. The benchmark in
docs/skills-architecture.mdshows the import-direct gain on a single-command read is negligible (~10ms); shell-out keeps the skill body simple and uses the same code path as anyone running the CLI manually. - The CLI is the source of truth for resolution logic — including override semantics (declarative wins over built-in for the same
(capability, provider)). Don't second-guess the registry's output here.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

