| name | asana-api |
| description | Call an arbitrary Asana REST API endpoint by HTTP method and path through the direct Asana API using a local ASANA_ACCESS_TOKEN. Use only as a low-level fallback when the normal Asana connector cannot reach the endpoint you need; you control the method, path, query, and body. |
| metadata | {"category":"productivity","capability_taxonomy_ids":["cap.productivity.task_tracking","cap.tools.api_automation"],"compatibility":{"runtimes":["codex","claude_code","cursor","antigravity","windsurf"]},"distribution_scope":"public","invocation_strategy":"explicit","version":"v0.1","source_of_truth":"https://github.com/nicshik/asana-skills"} |
Asana API
Use this skill as a generic passthrough to the Asana REST API when the normal
Asana connector or a higher-level skill cannot reach the endpoint or shape you
need. You choose the HTTP method, the path, the query parameters, and the
request body.
Prefer the official Asana connector and the focused skills for normal work.
This skill is a low-level tool: it can send writes, so use it deliberately.
Preconditions
ASANA_ACCESS_TOKEN is available in the shell, ASANA_ENV_FILE, --env-file, or a local .env.local.- The token belongs to an Asana user with access to the target resource.
- Network access to
https://app.asana.com/api/1.0 is allowed.
Non-Negotiable Rules
- Use
scripts/api.py.
- Never print the access token.
- Do not store the token in the skill directory.
- The method and path are sent verbatim; double-check before sending writes (
POST, PUT, DELETE).
- Pass query parameters with
--query k=v and body fields with --field k=v (both repeatable).
--field values are always sent as strings; for richer bodies use --data-file.--data-file wins over --field when both are given.- A
--data-file whose top-level JSON object already contains a data key is sent as-is; otherwise the loaded JSON is wrapped as the inner data object.
--all is honored only for GET and follows pagination via the shared client.
Recommended Commands
python3 scripts/api.py GET /tasks/1200000000000000 --opt-fields name,completed --json
python3 scripts/api.py GET /projects/1200000000000001/tasks --all --json
python3 scripts/api.py POST /tasks --field name=Example --field workspace=1200000000000000 --json
python3 scripts/api.py PUT /tasks/1200000000000000 --data-file ./body.json --json
Output Shape
- Text output:
rows=<n> for a list, gid= (and name= when present) for an object, or data= otherwise.
--json output: schema_version, fetched_at, and the data payload returned by the API.- Failure categories are
missing_token, bad_argument, not_found, permission_denied, premium_required, rate_limited, http_error, network, timeout, and api_error.
bad_argument means a --query, --field, or --data-file value was malformed; it is a local input problem, not a missing token.
