upsum
Close a work session by appending a durable record, rebuilding a fixed-size recency-weighted summary, carrying open work forward, and running repository checks.
Skill by lilCnid
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/lilcnid-upsum-cb1e0f
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/upsum
What's inside
- Skill name
upsum- Description
- Close a work session by appending a durable record, rebuilding a fixed-size recency-weighted summary, carrying open work forward, and running repository checks. Invoke explicitly.
- License
- CC-BY-4.0
- Instructions
- 95 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (4)
- LICENSE.md
- SKILL.md
- openai.yaml · agents/
- checks.py · scripts/
SKILL.md (first lines)
# Upsum
The record grows and never forgets. The summary stays bounded and forgets on
purpose. Run this ceremony only when explicitly invoked.
## Locate the project memory
Use existing repository conventions when present. Otherwise use:
| Path | Rule |
|---|---|
| `.upsum/RECORD.md` | append-only session record |
| `.upsum/SUMMARY.md` | rewritten whole, 250-word default budget |
| `TODO.md` | rewritten whole, human-visible open work |
## 1. Append one evidence-based record entry
Read the diff and repository state, then append:
```md
## {ISO_Date} · {Short_Session_Label}
**Did** — {Repository_Changes_Not_Conversation_Intent}
**Learned** — {New_Evidence_Or_Understanding}
**Decided** — {Decision_And_Owner}
**Left** — {Deliberately_Unfinished_Work_And_Reason}
```
Never edit earlier entries. If no work changed and no durable decision was made,
say so and stop instead of adding noise.
## 2. Rebuild the bounded summary
Derive `.upsum/SUMMARY.md` from the entire record. Keep the configured budget,
or 250 words when none exists. Allocate resolution by recency:
```text
oldest {Collapsed_To_A_Label}
{Outcome_Only}
{Outcome_Plus_One_Distinguishing_Detail}
newest {Nearly_Full_Detail_With_Redundancy_Trimmed}
```
Entries age by losing resolution, not by disappearing. Do not grow the budget
because the record grew.
## 3. Carry forward open work
Rewrite `TODO.md` from the record and current session. Each open item names the
blocking decision or condition and who can unblock it:
```md
- [ ] {Observable_Completion} — blocked on {Owner_And_Condition}
```
Remove completed items. `Left` records intentional incompletion; it is not an
apology field.
## 4. Run the shipped checks
Resolve this skill's directory from its source locator in the current skill
catalog, then run its script against the repository working directory:
```bash
python {upsum-skill-directory}/scripts/checks.py {repository-directory}
```
Do not assume the skill lives under a global config directory; plugins may run
from a repository scope or versioned cache. The script checks inside references,
repository state, skill health, and license/credit travel. `UNMEASURED` means the
check could not establish a result, not that it passed.
Exit codes describe measurement health:
- `0`: every check ran;
- `1`: at least one check could not be measured;
- `2`: the checker itself failed.
Act on findings or record why a finding is accepted. Do not merely narrate the
output.
## 5. Refresh publishing claims
Only when the session will commit, push, or release:
- test the documented install path from a clean location;
- verify that the README inventory matches the repository;
- verify that every stated limit still holds.
## Completion
Report which memory files changed, the summary budget, open blockers, check
results, and any unmeasured evidence. Preserve the append-only record exactly.From lilCnid/dovetail-codex, CC-BY-4.0.
More from lilCnid/dovetail-codex
- 0.526score
self-play
lilCnid
Evaluate prompts, skills, policies, or generated artifacts through clean-room Codex runs, adversarial cases, and evidence-backed revision.
SkillBFreshness: BClaude Code +9 - 0.526score
spark-steering
lilCnid
Diagnose a stuck or drifting Codex workflow by locating the wrong control surface, lifecycle, or owner before adding tools or permissions.
SkillBFreshness: BClaude Code +9 - 0.526score
subagent-composition
lilCnid
Partition a Codex task into bounded parallel investigations, choose delegated context deliberately, coordinate shared files, and consolidate evidence.
SkillBFreshness: BClaude Code +9 - 0.526score
judge-composition
lilCnid
Design and run reliable Codex judges with explicit rubrics, blinded evidence, calibration cases, and auditable verdicts.
SkillBFreshness: BClaude Code +9 - 0.526score
better-skill-creator
lilCnid
Create or revise Codex skills with precise routing, progressive disclosure, UI metadata, native validation, and behavior-focused evaluation.
SkillBFreshness: BClaude Code +9 - 0.516score
hypershot-protocol
lilCnid
Design prompt templates and model-facing artifacts with abstract structural examples that prime form without importing task-specific content.
SkillBFreshness: BClaude Code +9 - 0.496score
prompt-engineering
lilCnid
Structure or improve prompts, templates, agent instructions, and output schemas using clear hierarchy, semantic boundaries, placeholders, and attention control.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 3, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- CC-BY-4.0
- Repository
- github.com/lilCnid/dovetail-codex
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from lilCnid
- 0.574score
better-skill-creator
lilCnid
Use this skill for anything to do with authoring, fixing, or measuring Claude skills.
SkillBFreshness: B★ 4Claude Code +9 - 0.569score
subagent-composition
lilCnid
Compose a Claude Code sub-agent on the fly — persona, inherited-context transfer, tool budget, and return contract — as either an ephemeral Agent-tool call or…
SkillBFreshness: B★ 4Claude Code +9 - 0.569score
self-play
lilCnid
Run a clean-room self-play search over a problem space you have not solved yet — an AlphaGo-shaped loop over text where isolated sub-agent players (a blind…
SkillBFreshness: B★ 4Claude Code +9 - 0.569score
upsum
lilCnid
Close a working session: append what happened to a growing record, re-derive a fixed-length summary that holds the newest entries at full detail and collapses…
SkillBFreshness: B★ 4Claude Code +9 - 0.569score
judge-composition
lilCnid
Compose an adaptable four-role judge panel (grounding, coherence, corroboration, audit) to evaluate a candidate for promotion — a belief, claim, record entry…
SkillBFreshness: B★ 4Claude Code +9 - 0.564score
hypershot-protocol
lilCnid
Meta-prompting technique using zero-contamination structural examples.
SkillBFreshness: B★ 4Claude Code +9
Cite this page
upsum skill. Agent Mart. https://agentmart.dev/skills/lilcnid-upsum-cb1e0f (last updated 2026-10-24)