Commitment pattern miner (patterns over time) - #17
Merged
Conversation
First "patterns over time" layer — deterministic mining of the commitments table (no LLM): what's slipping (open + past due), where it concentrates (which people/kinds), and how often completed items finished late. Directly serves the "nothing slips" mission. - commitment_patterns(today): pure analyzer over existing columns (created_at/due_date/ completed_at/status/who/kind) → overdue list, late-completion rate, who/kind concentration, and a `findings` list of human sentences (only notable ones). - analyze_commitments tool: owner-only (NOT in the guest toolset); prompt steers "how am I doing / what keeps slipping" to it. Verified live: surfaces a real overdue item. - 2 tests (overdue concentration + late detection; clean slate → no findings). 227 pass. Audit note: this mines what's recorded today. Richer patterns (reschedule frequency, per-occurrence recurring on-time rate) would need a commitment_events log — deferred, since you can't mine history you didn't record; worth adding when we want those. Assisted-by: Claude Code (Opus 4.8)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
First "patterns over time" layer — and the high-signal one: deterministic mining of the commitments history (no fuzzy LLM inference), squarely on the "nothing slips" mission.
Audit result (what the DB records today)
commitmentshascreated_at,due_date,completed_at,status,who,kind,recurring. Enough for: overdue-open detection, late-completion rate, and who/kind concentration. Not enough (yet) for reschedule-frequency or per-occurrence recurring on-time rate — those need acommitment_eventslog (deferred; can't mine history you didn't record).What's here
commitment_patterns(today)— pure analyzer → overdue list (sorted, with days-overdue), late-completion rate, overdue-by-who / by-kind, and afindingslist of human sentences (only notable ones).analyze_commitmentstool — owner-only (correctly absent from the guest toolset); prompt steers "how am I doing / what keeps slipping" to it. Verified on live data — surfaced a real overdue item.Next (optional): a
commitment_eventslog to unlock reschedule/recurring-slip patterns; a proactive weekly surfacing; and the softer weekly-reflection-over-memory layer.🤖 Generated with Claude Code