.md file to compare - side-by-side diff against catchup
catchup
description: "Triggers on prompt mention of 'catchup'."
What it does for you
Gets you ready for a client meeting with what's happened since you last spoke.
What it produces
A recent result, so you can see the kind of work it returns.
loading…
How to get it
These run inside the Snappy workspace. Want this working in your business? I set skills like this up with you, in one focused week.
For developers how this skill is built, graded, and how it runs
at a glance- the short version
what's inside - the parts that make up a skill 2/4 present
A skill is just a few plain-text files. Only the main one is required. The rest are optional, added as the work needs them. This is what the skill is made of; how it runs is just below.
state/skills/catchup/SKILL.md
present
state/lib/catchup.ts
not present
state/bin/catchup/
not present
state/skills/catchup/AGENTS.md
present
how it's graded - what counts as a good run 4 criteria · 3 deterministic · 1 judge
Each row is one thing a good run has to get right. deterministic means a quick check decides, pass or fail. judge means the AI reads the result and rates it. Grading each piece on its own (instead of one overall score) shows exactly where a run fell short, so the fix is obvious.
how it runs - the shared frame every skill uses 3/5 present
Every skill runs the same way. One part does the work, a separate part checks it, and a short loader hands the AI exactly what it needs for the job. Anything this skill doesn't use shows a one-line note saying why, on purpose, not by accident.
state/log/evals.ndjson - ALWAYS run voice.checkTone() on the dispatched draft — internal briefs get pasted into client comms, so em-dash check still applies (score 0.0 on tone fail)
- ALWAYS resolve since from ~/.claude/cache/krisp/meetings.json last-meeting timestamp UNLESS explicitly overridden — Krisp can mislabel meetings, so allow override
- Client repo path convention is hard-coded: ~/projects/snappy-client-<slug> — don't guess slugs
- All four signals (commits, messages, commitments, invoice_state) populated → 1.0; some genuinely empty → 0.5; tone fail → 0.0
what it has learned - fixes written back in over time sample
When a run hits something this skill didn't handle, the fix gets written back into the skill so it doesn't happen again. FIXED means it was corrected on the spot. LOGGED means it's queued for a bigger rewrite. Either way, the skill gets a little better and never makes the same mistake twice.
- Loading feedback rows…
how the work flows- step by step
`client` is required input; collects 4 signals in parallel, dispatches `gemini` to condense, gates on `voice.checkTone()`, writes `state/log/catchup/<client>-<date>.md`
SKILL.md- the skill, written out in plain English
Backed by: state/lib/sweep.ts + state/lib/knowledge.ts
catchup
One-screen brief before a meeting. Pulls:
git log --since=$sincein~/projects/snappy-client-<slug>- Sweep items filtered to client participants
- Krisp action items owned-by-Robert for the client
- Xano invoice state
Steps
- Resolve
sincefrom~/.claude/cache/krisp/meetings.json(last
meeting with this client) if not passed.
- Collect the four signals.
- Dispatch
geminito condense into a one-screen markdown brief. voice.checkTone()gate (internal drafts are §4a-exempt for
frontmatter but still must pass em-dash check).
- Write to
state/log/catchup/<client>-<date>.md.
Eval
score("catchup", run_id, {
score:
shape_ok && tone_ok && has_all_four_signals ? 1.0 :
shape_ok && tone_ok ? 0.5 :
0.0,
signals_present: {
commits: commits.length > 0,
messages: msgs.length > 0,
commitments: commits_k.length > 0,
invoice_state: !!invoice,
},
primary_issue: !tone_ok ? "voice-gate-failed" : null,
});
0.5 acceptable if a signal genuinely has no data (new client, no commits yet). 1.0 when all four populate. 0.0 on tone fail - even internal drafts go through the gate because they get pasted into client comms.
Gotchas
sinceresolution from Krisp can be wrong if the last meeting was
mislabeled. Allow explicit override.
- Client repo path convention:
~/projects/snappy-client-<slug>.
Rubric
criteria:
- name: all_four_signals_present
kind: deterministic
check: "The 'signals_present' field in the evaluation JSON reports 'true' for 'commits', 'messages', 'commitments', and 'invoice_state'."
- name: brief_tone_passes_gate
kind: deterministic
check: "The 'primary_issue' field in the evaluation JSON is 'null' or does not contain 'voice-gate-failed'."
- name: output_file_created
kind: deterministic
check: "A markdown file named 'catchup-<client>-<date>.md' exists in 'state/log/catchup/'."
- name: brief_content_relevance
kind: judge
check: "The generated brief accurately condenses and presents information from the four source signals (git log, sweep items, Krisp items, Xano invoice state) for the specified client."AGENTS.md- what the AI loads when this skill comes up
catchup - loader
Per-turn rules for the catchup skill. Full reference: state/skills/catchup/SKILL.md. Do not skip these.
Critical Rules
- ALWAYS run
voice.checkTone()on the dispatched draft - internal briefs get pasted into client comms, so em-dash check still applies (score 0.0 on tone fail) - ALWAYS resolve
sincefrom~/.claude/cache/krisp/meetings.jsonlast-meeting timestamp UNLESS explicitly overridden - Krisp can mislabel meetings, so allow override - Client repo path convention is hard-coded:
~/projects/snappy-client-<slug>- don't guess slugs - All four signals (commits, messages, commitments, invoice_state) populated → 1.0; some genuinely empty → 0.5; tone fail → 0.0
Commands
| ui dashboard | state/skills/catchup/resources/ui.openui | |invoke: client is required input; collects 4 signals in parallel, dispatches gemini to condense, gates on voice.checkTone(), writes state/log/catchup/<client>-<date>.md |verify: shape_ok && tone_ok && has_all_four_signals |eval log: state/log/evals.ndjson (skill: "catchup")
OpenUI Resource
- Skill-owned OpenUI Lang resource:
state/skills/catchup/resources/ui.openui. Read it before rendering or editing this skill's generated component surface. - Treat this resource as a first-class artifact of the skill, not a generic chat response. Improve it when the skill's user-facing output needs to become richer.
- System resources compose OpenUI primitives and inherit SnappyChat tokens. Use
ui_contract: brandedin SKILL.md only for deliberate platform or client visuals.
Known Pitfalls
sinceresolution can be wrong if the last Krisp meeting was mislabeled - allow explicit override via thesinceinput- New clients legitimately empty in commits/commitments - that's a 0.5, not a 0.0
- Frontmatter is §4a-exempt for image-prompt fields, but body prose is not (memory: feedback_image_prompts_not_4a)
Self-Test
An agent reading this should correctly:
- [ ] Reject a draft that fails
checkTone()even though it's an internal brief - [ ] Use
~/projects/snappy-client-<slug>for repo path - [ ] Score 0.5 (not 0.0) when a new client has no commits yet
Self-report
If this loader fell short, append a line:
echo "[$(date -u +%FT%TZ)] catchup: <what was missing>" >> state/log/loader-feedback.log
<!-- kernel-ok: historical attribution for the PID-loader pattern. -->
Self-correcting loader (PID feedback)
This is the rule that makes loaders learn. Ported from snappy-kernel, where it demonstrably worked: the snappy-dom-cartographer skill mapped Skool admin UI once, every later ship-agent inherited the full map for free.
Before you finish, do two things:
1. Fix gaps (P - proportional). MANDATORY. If this loader didn't cover your case - if you had to read another file, run an undocumented command, work around a wrong selector, or discover a quirk - you MUST attempt an Edit to this AGENTS.md before you log.
- Default action: edit this file. Surgical. Add the missing line, table
row, or code example where it belongs. No restructuring.
- Only
LOGGEDis allowed when: the fix needs >10 lines, spans multiple
files, or requires a structural rewrite. In that case the state/regen/drain.sh queue picks it up asynchronously.
- "I didn't have time" / "it's minor" / "the next agent will figure it out"
are NOT valid reasons. Minor gaps compound - that's the whole point of this loop.
- Concrete example of a valid P-fix: discover
agent-browser screenshot
takes path positionally not as --output <path>. One line in a table. Edit it in. Do not just log it.
- The goal: the next agent never has to leave the loader.
2. Log the result.
echo "[$(date -u +%FT%TZ)] catchup: <what was missing or fixed> [FIXED|LOGGED]" >> state/log/loader-feedback.log
FIXED= you patched this loader inline (P-fix).LOGGED= too large for inline; the PostToolUse enqueue + Stop-hook drain
will rewrite the loader from scratch on next session-end.
Do not skip this. Every agent run must leave the system better than it found it. The loader is the setpoint; you are the sensor; the gap is the error signal; closing the gap is the correction.
api.ts- the code it can call
⚠ no api.ts - this skill has no typed action surface
scripts- helper scripts it can run
prose-only skill - 2 inline code blocks live in SKILL.md above (no state/bin/ sidecar yet).
how we check it- the checks, plus the last 10 runs
| timestamp | verb | score | primary_issue | artifact |
|---|---|---|---|---|
| 2026-04-25 04:11Z | - | 1.00 | - | - |
| 2026-04-21 15:58Z | - | 1.00 | - | - |
| 2026-04-21 15:56Z | - | 1.00 | - | - |
| 2026-04-21 03:53Z | - | 1.00 | - | - |
| 2026-04-25 04:11Z | - | 1.00 | - | - |
| 2026-04-21 15:58Z | - | 1.00 | - | - |
| 2026-04-21 15:56Z | - | 1.00 | - | - |
| 2026-04-21 03:53Z | - | 1.00 | - | - |
| 2026-04-25 04:11Z | - | 1.00 | - | - |
| 2026-04-21 15:58Z | - | 1.00 | - | - |