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 Knuckles-Team/gitlab-api --skill gitlab-api-developmentgit clone --depth 1 https://github.com/Knuckles-Team/gitlab-apiWrote 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/knuckles-team/gitlab-api/gitlab-api-development)<a href="https://agentmods.dev/skills/knuckles-team/gitlab-api/gitlab-api-development"><img src="https://agentmods.dev/badge/skills/knuckles-team/gitlab-api/gitlab-api-development/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/knuckles-team/gitlab-api/gitlab-api-development"><img src="https://agentmods.dev/badge/skills/knuckles-team/gitlab-api/gitlab-api-development.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 | $0.00049 | $0.01515 |
| Opus 5.5 | $0.00020 | $0.00606 |
| Sonnet 5.5 | $0.00010 | $0.00303 |
| Haiku 4.5 | $0.00005 | $0.00152 |
Grade A, and why
gitlab-api-development 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 yesterday.
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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
gitlab-api development
This repository owns the GitLab REST and GraphQL API facades, tenant and
authentication boundary, action-routed MCP tools, GitLab ontology and KG
ingestion mappings, and packaged connector content. Shared connector
contracts, transports, TLS profiles, and credential helpers belong to
agent-connector-sdk; admission, serving, and durable graph state belong to
their owning repositories. Read AGENTS.md before changing code and keep
credentials, private endpoints, tenant data, and customer records out of
source, logs, fixtures, and skill content.
Rapid delivery: the 5-minute contract
Specs are already designed. Deliver small, reviewable slices and cite the owning spec row in the commit trailer.
- Find the owning row first with
git grep -n "<term>" -- specsand extend the existing spec before adding requirement text. - Reuse the single
Connectoringitlab_api/__init__.py, theApifacade ingitlab_api/api_client.py, the modular clients undergitlab_api/api/, and the action dispatch tables ingitlab_api/mcp_server.py. Keep GraphQL operations ingitlab_api/gitlab_gql.py; do not create a second API, MCP, authentication, or transport plane. - Preserve the multi-tenant resolution in
gitlab_api/instances.py. A named instance, bare URL, or default selection must retain caller entitlement checks and theGITLAB_INSTANCESshared setting. The legacyGITLAB_URL/GITLAB_TOKENfallback remains supported. - Preserve
GITLAB_URL,GITLAB_TOKEN,GITLAB_TLS_PROFILE, andENABLE_DELEGATIONhandling ingitlab_api/auth.py. Delegated RFC 8693 token exchange and fixed credentials stay behind the SDK identity/TLS helpers. Never print tokens or add a TLS verification bypass. - Keep the REST resource paths, GraphQL typed-operation boundary, action names,
pagination, and response models stable. Use
run_blockingfor blocking API calls from MCP tools and return typed refusal/error envelopes instead of bypassing the shared dispatcher. KeepMCP_TOOL_MODE(condensed,verbose, orboth) and each action group's toggle behavior aligned with the generated tool surface. - Preserve typed KG ingestion through
gitlab_api/kg_ingest.py, source declarations ingitlab_api/connector_sources.py, and neutral ontology mappings. Project, issue, merge-request, pipeline, job, runner, group, and person records use the existing stable IDs and relationship names. - Mock HTTP, GraphQL, and authentication responses in local tests; never call a
live GitLab instance or use live credentials in a gate. Every delivered
requirement needs a bound focused test. Generated
specs/*/status.jsonfiles come from shared hooks; never hand-edit them. - Keep
skills/**,ontology/**,connectors/**, prompts, andmcp_config.jsoncoherent with package data and theagent_connector_sdk.connectorentry point (gitlab-mcp = gitlab_api:connector). Do not add a second publisher or runtime plane.
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.
- yesterday First seen · 133 lines · 49 tokens per session scan A 441fc79d0191
gitlab-api-development is a skill published in the GitHub repository Knuckles-Team/gitlab-api (10 stars, last pushed yesterday), licensed MIT. It adds 49 tokens to every session and 1,515 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-27.
Other skills, from other repositories
api-connector-builder
Use when writing a client for someone else's REST or GraphQL API: auth flow choice and token refresh, pagination to exhaustion, retry-with-jitter on transient failures only, rate-limit-aware throttling. NOT inbound callbacks (that is webhooks), NOT chaining services (that is automation-flows), NOT designing your own…
freshrss-agent-development
Develop, test, gate, and release the freshrss-agent connector. Use this skill when changing FreshRSS Reader API mappings, governed MCP tools, delegated or fixed authentication, KG ingestion, packaged content, or repository gates.
api-integration-architect
Design, implement, debug, and optimize API integrations with expert-level patterns for REST, GraphQL, webhooks, and authentication flows.
api-testing-patterns
Comprehensive API testing patterns including contract testing, REST/GraphQL testing, and integration testing. Use when testing APIs or designing API test strategies.
api-design-patterns
Comprehensive API design patterns covering REST, GraphQL, gRPC, versioning, authentication, and modern API best practices.
pipefy-api-fallback
Use this skill when a toolkit operation fails AND the introspection skill could not resolve the problem. This is the last-resort fallback (Tier 3): call the Pipefy GraphQL API directly using curl or httpx, authenticating with the Service Account (OAuth2) or a Personal Access Token (PAT) available as env var. Follow…