← ClaudeAtlas

backloglisted

Answers what to work on next and where a new request fits, against whichever ticket system the project uses or against the FAW internal registry. Use at the start of a working session or when what comes next is unclear.
DiLoretoT/faw · ★ 0 · AI & Automation · score 68
Install: claude install-skill DiLoretoT/faw
# Backlog Two questions: **what do I work on next?** and **where does this request fit?** ## Where the tickets come from Declared by `tickets.system` in `.faw/config.json`. If the file does not exist, the value is `internal`. | System | Where they are read | How they are operated | |---|---|---| | `internal` | `docs/faw/tickets/*.md` in the repository | Read and edited as files; git provides the history | | `ado`, `jira`, `github` | The declared tracker | Through its MCP server if connected in the session; if not, the user operates the tool and the work happens with whatever they report | | `none` | There is no backlog | Work on what the user asks for at the time | **Before reading an external tracker, check whether an MCP server is available for it.** With MCP, read and update it directly. Without MCP, do not invent the state of the backlog and do not assume a ticket exists. Ask the user, or work with the identifier they give. A tool with no MCP server does not prevent using FAW. The only thing the method needs from a ticket is its identifier. ## Mode: what do I work on next 1. **Get the real state.** From the declared tracker or the internal registry, not from memory or from what was discussed last session. 2. **Lay out the picture, short:** ``` Open : <N> In progress: <id> <title> Blocked : <id> <title> - why Next : <id> <title> <- proposal ``` 3. **Propose the next one ordered by technical dependency, not by backlog order.*