.config/opencode/skills/tecton/SKILL.md
Run Tecton plan and tests via Pants in the data-science repo. Handles long-running commands with proper output capture to avoid truncation.
npx skillsauth add ferdinandyb/dotfiles tectonInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Tecton commands (especially tecton-plan) produce very long output that gets truncated.
Always use temp file capture pattern:
TMPFILE=$(mktemp)
PANTS_CONCURRENT=True pants tecton-plan --workspace=<workspace> --skip-tests 2>&1 | tee "$TMPFILE"; echo "EXIT_CODE: $?"
# Inspect results
tail -200 "$TMPFILE"
# If needed, inspect more or search for specific errors
rg "error|Error|ERROR" "$TMPFILE"
# Clean up when done
rm -f "$TMPFILE"
Key points:
tee to capture output to temp file while still streamingecho "EXIT_CODE: $?"tail or rg to inspect the temp file as neededPANTS_CONCURRENT=True when running multiple pants invocations concurrentlycheckoutandsale-v1-staging: Staging workspace - check against this firstbence-dev: Personal dev workspace for clarifying issues if staging fails with confusing errorsIf staging passes, no need to check against bence-dev. Use bence-dev only when staging errors are unclear.
Runs tecton plan against a workspace. Use --skip-tests to speed up iteration.
# Plan against staging (check this first)
TMPFILE=$(mktemp)
PANTS_CONCURRENT=True pants tecton-plan --workspace=checkoutandsale-v1-staging --skip-tests 2>&1 | tee "$TMPFILE"; echo "EXIT_CODE: $?"
tail -200 "$TMPFILE"
rm -f "$TMPFILE"
# Plan against dev workspace (only if staging errors are unclear)
TMPFILE=$(mktemp)
PANTS_CONCURRENT=True pants tecton-plan --workspace=bence-dev --skip-tests 2>&1 | tee "$TMPFILE"; echo "EXIT_CODE: $?"
tail -200 "$TMPFILE"
rm -f "$TMPFILE"
Run Tecton tests using the pants test goal with --tecton-test-run flag.
# Run all tecton tests in a directory
pants test --tecton-test-run projects/features/repos/sig_tecton/tests/sig_tecton/datasources/orders/::
# Run a specific test file
pants test --tecton-test-run projects/features/repos/sig_tecton/tests/sig_tecton/datasources/orders/test_orders_datasources.py
# Run a specific test function
pants test --tecton-test-run projects/features/repos/sig_tecton/tests/path/to/test_file.py -- -k test_function_name
For long test runs, use the same temp file capture pattern:
TMPFILE=$(mktemp)
pants test --tecton-test-run projects/features/repos/sig_tecton/tests/:: 2>&1 | tee "$TMPFILE"; echo "EXIT_CODE: $?"
tail -200 "$TMPFILE"
rm -f "$TMPFILE"
documentation
IMMEDIATELY load this skill if you read ANY file containing "taskagent" (PLAN.md, org/projects, docs, etc). Also load for multi-session work or when user mentions agent tasks. For simple single-session tasks without taskagent references, use TodoWrite instead.
testing
Load this skill when performing git operations that modify repository state: add, commit, amend, rebase, merge, stash, worktree, reset, checkout, switch, branch create/delete.
tools
MUST load before ANY git or yadm operation — reading, writing, branching, committing, diffing, logging, rebasing, or anything else git-related. No exceptions.
testing
Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md file. Also use when editing or restructuring a skill directory (moving files to references/ or scripts/, removing stale content, validating against the AgentSkills spec). Triggers on phrases like "create a skill", "author a skill", "tidy up a skill", "improve this skill", "review the skill", "clean up the skill", "audit the skill".