Skill metadata
Reference: full SKILL.md
The following is the complete skill definition that Mibyan loads when this skill is triggered. This is what the agent sees as instructions when the skill is active.
Airtable — Bases, Tables & Records
Work with Airtable’s REST API directly viacurl using the terminal tool. No MCP server, no OAuth flow, no Python SDK — just curl and a personal access token.
Prerequisites
- Create a Personal Access Token (PAT) at https://airtable.com/create/tokens (tokens start with
pat...). - Grant these scopes (minimum):
data.records:read— read rowsdata.records:write— create / update / delete rowsschema.bases:read— list bases and tables
- Important: in the same token UI, add each base you want to access to the token’s Access list. PATs are scoped per-base — a valid token on the wrong base returns
403. - Store the token in
${mibyan_HOME:-~/.mibyan}/.env(or viamibyan setup):
Note: legacy key... API keys were deprecated Feb 2024. Only PATs and OAuth tokens work now.
API Basics
- Endpoint:
https://api.airtable.com/v0 - Auth header:
Authorization: Bearer $AIRTABLE_API_KEY - All requests use JSON (
Content-Type: application/jsonfor any POST/PATCH/PUT body). - Object IDs: bases
app..., tablestbl..., recordsrec..., fieldsfld.... IDs never change; names can. Prefer IDs in automations. - Rate limit: 5 requests/sec/base.
429→ back off. Burst on a single base will be throttled.
-s suppresses curl’s progress bar — keep it set for every call so the tool output stays clean for Mibyan. Pipe through python -m json.tool (always present) or jq (if installed) for readable JSON.
Field Types (request body shapes)
Pass
"typecast": true at the top level of a create/update body to let Airtable auto-coerce values (e.g. create a new select option on the fly, convert "42" → 42).
Common Queries
List bases the token can see
List tables + schema for a base
options.choices for select fields, and shows primary-field names.
List records (first 10)
Get a single record
Filter records (filterByFormula)
Airtable formulas must be URL-encoded. Let Python stdlib do it — never hand-encode:- Exact match:
{Email}='user@example.com' - Contains:
FIND('bug', LOWER({Title})) - Multiple conditions:
AND({Status}='Todo', {Priority}='High') - Or:
OR({Owner}='alice', {Owner}='bob') - Not empty:
NOT({Assignee}='') - Date comparison:
IS_AFTER({Due}, TODAY())
Sort + select specific fields
%5B / %5D).
Use a named view
Common Mutations
Create a record
Create up to 10 records in one call
Update a record (PATCH — merges, preserves unchanged fields)
Upsert by a merge field (no ID needed)
performUpsert creates records whose merge-field values are new, patches records whose merge-field values already exist. Great for idempotent syncs.
Delete a record
Delete up to 10 records in one call
Pagination
List endpoints return at most 100 records per page. If the response includes"offset": "...", pass it back on the next call. Loop until the field is absent:
Typical Mibyan Workflow
- Confirm auth.
curl -s -o /dev/null -w "%{http_code}\n" https://api.airtable.com/v0/meta/bases -H "Authorization: Bearer $AIRTABLE_API_KEY"— expect200. - Find the base. List bases (step above) OR ask the user for the
app...ID directly if the token lacksschema.bases:read. - Inspect the schema.
GET /v0/meta/bases/$BASE_ID/tables— cache the exact field names and primary-field name locally in the session before mutating anything. - Read before you write. For “update X where Y”,
filterByFormulafirst to resolve therec...ID, thenPATCH /v0/$BASE_ID/$TABLE/$RECORD_ID. Never guess record IDs. - Batch writes. Combine related creates into one 10-record POST to stay under the 5 req/sec budget.
- Destructive ops. Deletions can’t be undone via API. If the user says “delete all Xs”, echo back the filter + record count and confirm before firing.
Pitfalls
filterByFormulaMUST be URL-encoded. Field names with spaces or non-ASCII also need encoding ({My Field}→%7BMy%20Field%7D). Use Python stdlib (pattern above) — never hand-escape.- Empty fields are omitted from responses. A missing
"Assignee"key doesn’t mean the field doesn’t exist — it means this record’s value is empty. Check the schema (step 3) before concluding a field is missing. - PATCH vs PUT.
PATCHmerges supplied fields into the record.PUTreplaces the record entirely and clears any field you didn’t include. Default toPATCH. - Single-select options must exist. Writing
"Status": "Shipping"whenShippingisn’t in the field’s option list errors withINVALID_MULTIPLE_CHOICE_OPTIONSunless you pass"typecast": true(which auto-creates the option). - Per-base token scoping. A
403on one base while another works means the token’s Access list doesn’t include that base — not a scope or auth issue. Send the user to https://airtable.com/create/tokens to grant it. - Rate limits are per base, not per token. 5 req/sec on
baseAand 5 req/sec onbaseBis fine; 6 req/sec onbaseAalone will throttle. Monitor theRetry-Afterheader on429.
Important Notes for Mibyan
- Always use the
terminaltool withcurl. Do NOT useweb_extract(it can’t send auth headers) orbrowser_navigate(needs UI auth and is slow). AIRTABLE_API_KEYflows from${mibyan_HOME:-~/.mibyan}/.envinto the subprocess automatically when this skill is loaded — no need to re-export it before eachcurlcall.- Escape curly braces in formulas carefully. In a heredoc body,
{Status}is literal. In a shell argument,{Status}is safe outside{...}brace-expansion context — but pass dynamic strings throughpython urllib.parse.quotebefore splicing into a URL. - Pretty-print with
python -m json.tool(always present) rather thanjq(optional). Only reach forjqwhen you need filtering/projection. - Pagination is per-page, not global. Airtable’s 100-record cap is a hard limit; there is no way to bump it. Loop with
offsetuntil the field is absent. - Read the
errorsarray on non-2xx responses — Airtable returns structured error codes likeAUTHENTICATION_REQUIRED,INVALID_PERMISSIONS,MODEL_ID_NOT_FOUND,INVALID_MULTIPLE_CHOICE_OPTIONSthat tell you exactly what’s wrong.

