td-ai-tools 1.0.2

package/skills/car-ticket-generator/SKILL.md

@@ -0,0 +1,130 @@
+---
+name: car-ticket-generator
+description: Generate a ticket for the codex-auto-runner queue
+alwaysApply: false
+---
+
+# CAR Ticket Skill
+
+Generate a ticket for the Codex-Auto-Runner ticket queue
+
+---
+
+## 1) Ticket files
+
+- Filename: `TICKET-###*.md`
+  Examples: `TICKET-001.md`, `TICKET-120-api-parity.md`
+- Tickets **must be in ascending numeric order**.
+- Numbers **do not need to be consecutive**.
+  Leave gaps if follow-up tickets are likely.
+- Location: `.codex-autorunner/tickets`
+
+---
+
+## 2) Required frontmatter (minimum)
+
+```yaml
+---
+agent: 'codex'
+done: false
+---
+```
+
+Required (linted):
+
+- `agent`: registered CAR agent id for this repo (for example `codex`, `opencode`) or the special value `user`
+- `done`: boolean
+
+Do not use assistant product names as `agent` values (for example `chatgpt` or `claude`) unless those exact ids are configured in CAR for this repo.
+
+Common optional fields:
+
+- `title`: short, outcome-focused summary
+- `goal`: one-sentence statement of intent
+- `model`: pin a specific model when necessary
+
+---
+
+## 3) Recommended ticket body structure
+
+Keep tickets concise and independently verifiable.
+
+Preferred sections:
+
+- `## Tasks` - concrete implementation steps
+- `## Acceptance criteria` (or `## Exit criteria`) - observable outcomes
+- `## Tests` - commands or explicit verification steps
+- `## Notes` - only if necessary
+
+Write criteria so another agent can prove completion without guesswork.
+
+---
+
+## 4) Sequencing rules (critical)
+
+CAR's `ticket_flow`:
+
+- Executes tickets in ascending order
+- Picks the first ticket where `done != true`
+
+Implications:
+
+- Put prerequisites in lower-numbered tickets.
+- Never make a lower-numbered ticket depend on a higher-numbered one.
+- If reverse dependencies appear, reorder or split tickets.
+- Each ticket must be independently completable when its turn arrives.
+
+---
+
+## 5) Assignment defaults
+
+- Implementation work -> repo agents (`codex`, `opencode`, etc.).
+- Final human review/signoff -> user-assigned ticket near the end.
+- In PMA mode, prefer delegation, not direct code edits.
+
+---
+
+## 6) Quality bar for good tickets
+
+A good ticket:
+
+- Has a single, well-scoped outcome
+- References specific files/modules when useful
+- Includes explicit verification (tests, checks, or observable behavior)
+- Avoids vague language (`"improve"`, `"clean up"`, `"fix issues"`) without criteria
+
+---
+
+## 7) Copy-paste ticket template
+
+```md
+---
+title: '<Outcome-focused title>'
+agent: 'codex'
+done: false
+goal: '<What will be true when this ticket is complete>.'
+---
+
+## Tasks
+
+- <Concrete implementation step>
+- <Concrete implementation step>
+
+## Acceptance criteria
+
+- <Observable behavior or artifact>
+- <Observable behavior or artifact>
+
+## Tests
+
+- <Commands to run or explicit checks>
+```
+
+---
+
+## 8) Anti-patterns to reject
+
+- Missing or invalid frontmatter
+- Cross-ticket dependency deadlocks
+- Tickets that rely on unstated or hidden context
+- `done: true` without evidence against acceptance criteria

package/skills/car-ticket-generator/agents/openai.yaml

@@ -0,0 +1,4 @@
+interface:
+    display_name: 'Car Ticket Generator'
+    short_description: 'Generate a car ticket for the user'
+    default_prompt: 'Generate a car ticket for the user'

package/skills/everhour-basecamp-estimates/.env.example

@@ -0,0 +1,2 @@
+EVERHOUR_API_KEY=replace-with-everhour-api-key
+EVERHOUR_PROJECT_ID=ev:1234567890

package/skills/everhour-basecamp-estimates/SKILL.md

@@ -0,0 +1,73 @@
+---
+name: everhour-basecamp-estimates
+description: Bulk update Everhour task estimates from a Basecamp todo or todolist URL, then append bracketed hours to the Basecamp todo titles.
+alwaysApply: false
+---
+
+Use this skill when the user wants to update Everhour task estimates in bulk from a Basecamp todo or todolist URL.
+
+## Files
+
+- Local config: `.agents/skills/everhour-basecamp-estimates/.env`
+- Example config: `.agents/skills/everhour-basecamp-estimates/.env.example`
+- Workflow script: `.agents/skills/everhour-basecamp-estimates/scripts/update_estimates.py`
+
+## Required inputs
+
+- A Basecamp todo or todolist URL
+- Either one estimate in hours to apply to every active task, or a JSON array of per-task estimates in hours
+
+## Workflow
+
+1. Confirm `.agents/skills/everhour-basecamp-estimates/.env` exists and contains:
+   - `EVERHOUR_API_KEY`
+   - `EVERHOUR_PROJECT_ID`
+2. Run the bundled script from the repo root:
+
+```bash
+python3 .agents/skills/everhour-basecamp-estimates/scripts/update_estimates.py \
+  "https://3.basecamp.com/..." \
+  2.5
+```
+
+3. For todolists that need different estimates per todo, pass a JSON array. The array length must match the number of active todos after completed items are skipped:
+
+```bash
+python3 .agents/skills/everhour-basecamp-estimates/scripts/update_estimates.py \
+  "https://3.basecamp.com/..." \
+  "[1,2.5,4]"
+```
+
+4. Use `--dry-run` first if the user wants to review matches before mutating anything:
+
+```bash
+python3 .agents/skills/everhour-basecamp-estimates/scripts/update_estimates.py \
+  "https://3.basecamp.com/..." \
+  "[1,2.5,4]" \
+  --dry-run
+```
+
+## Behavior
+
+- The script parses the Basecamp URL with the Basecamp CLI.
+- For a todo URL, it loads one todo title.
+- For a todolist URL, it loads every todo title in that list.
+- Completed Basecamp todos are skipped.
+- A scalar estimate is applied to each active todo.
+- A JSON estimate array is applied in active-todo order and must match the active todo count exactly.
+- It matches Basecamp titles against Everhour task names inside the configured project.
+- Matching is exact after normalization:
+  - trim whitespace
+  - collapse repeated spaces
+  - case-insensitive compare
+  - strip a trailing estimate suffix like `[2.5h]` before matching
+- It updates Everhour estimates first, converting hours to seconds and sending an overall estimate.
+- Only after all Everhour updates succeed does it rename the Basecamp todos to append the bracketed hours suffix.
+
+## Guardrails
+
+- If any Basecamp title has no Everhour match, stop and report it.
+- If any Basecamp title matches more than one Everhour task, stop and report the ambiguity.
+- If an estimate array length does not match the active todo count, stop before any updates.
+- If the provided todo URL is already completed, stop without updating anything.
+- Do not update Basecamp titles before Everhour updates complete.

package/skills/everhour-basecamp-estimates/agents/openai.yaml

@@ -0,0 +1,4 @@
+interface:
+    display_name: 'Everhour Estimates'
+    short_description: 'Bulk sync Everhour estimates from Basecamp todos.'
+    default_prompt: 'Use $everhour-basecamp-estimates to update Everhour task estimates from a Basecamp todo or todolist URL.'