Deeds

A New Way to Measure Work Done with AI

Pull requests show that code changed. Deeds tries to show what got better.

$curl -fsSL https://raw.githubusercontent.com/danielmiessler/deeds/main/install.sh | sh

Runs on your machine. All you need is a Cloudflare account. Source on GitHub

deeds analyze github.com/golang/go --since 90d --html report.html
Open the full report
1,065 commits to Go in 90 days. 1,041 of them did something deeds counts. 146 caps494 fixes887 tends

I'm not sure we're measuring AI coding work correctly.

The pull request has been the unit of software work for a long time, and it has earned that. It gives a team one clean place to discuss a change and review it before it merges. I'm not trying to replace it for that.

But as a way to measure how much work got done, especially work done with AI, I see two problems.

  1. It's built for collaboration.

    A PR exists so someone else can review your change. One person working with agents, testing and pushing straight to production, often never opens one. The share of code written that way is rising fast.

  2. It shows change, not value.

    A PR tells you code changed. A refactor, a typo fix and a revert count the same as a new capability, and a change that made things worse counts too.

As PRsrefactorreverttoken leakCSV export4 PRs
As deedstendno deedfixcap1 cap, 1 fix, 1 tend

I don't think the PR is the right unit of measurement anymore.

Deeds is my attempt to address both problems. It's a proposal, not a finished answer, and I expect parts of it to be wrong. If you see where, I'd like to hear it.

Go's last 90 days, counted two ways.

Counting activity

1,065

commits, and zero GitHub pull requests: Go reviews its changes in Gerrit. A count says a lot happened. It can't say what.

Counting deeds

146
caps
494
fixes
887
tends

146 things Go can do that it couldn't, or does more fully. 494 things that work now. And the upkeep that kept it standing.

The proposal: deeds.

Deeds is a proposed alternative for measuring valuable work rather than collaborative code modifications. It works whether or not anyone opens a PR, and it only counts a change when the product got better. Each of those changes is a deed, and every deed is one of three kinds.

cap

A capability gained or deepened. Someone can now do a thing they couldn't, or do an existing thing more fully.

src/reports/export.ts
+ export function toCSV(rows: Invoice[]): string {
+   return [HEADER, ...rows.map(toLine)].join("\n");
+ }
fix

Something moved from broken to sound. Security fixes count here too.

src/server/session.ts
- log.info("session", req.headers.cookie);
+ log.info("session", redact(req.headers.cookie));
tend

Upkeep nobody sees: refactors, dependencies, performance, tests.

package.json
-   "hono": "^4.5.1",
+   "hono": "^4.6.0",

Judged from the diff

Every commit is read from what it actually changed, never from its message. Renames, reverts and busywork add no caps.

Counts stay separate

Nine caps, four fixes and six tends tells a story one blended score would hide.

No ranking of domains

A recipe app and a payments system meet on the same ground. A team that wants one number can apply its own weights on top.

Point it at any repo.

  1. Pick a repo and a windowA local folder or a GitHub URL, over the last week, a quarter, or all time.
  2. Every commit is judgedEach diff becomes zero or more caps, fixes and tends.
  3. You get the reportTotals, a week-by-week trend, and who did the work. Same repo and window, same answer.
deeds analyze github.com/golang/go --since 90d
deeds github.com/golang/go
2026-07-07 → 2026-10-05 · 1065 commits (1064 cached) · judged by clef (clef)
​
146 caps 494 fixes 887 tends
​
week of caps fixes tends
07-06 ▌ 1 ██▍ 11 ██▍ 19
07-13 ██████ 13 ████████████ 57 ███████████▌ 91
07-20 ██▊ 6 █████▋ 27 ██████▏ 48
07-27 ██▊ 6 ████▎ 20 ████▏ 33
08-03 ██▎ 5 █████ 24 █████▍ 43
08-10 ███▊ 8 ████████▎ 39 █████████▋ 76
08-17 ███████▍ 16 ████████▎ 39 ██████████▋ 84
08-24 ██████▌ 14 ███████████▍ 54 ███████████ 87
08-31 ████████████ 26 ████████▎ 39 ██████████▋ 84
09-07 ██▊ 6 █████████ 43 ████████▌ 67
09-14 ███▎ 7 ████████ 38 █████▌ 44
09-21 ████▏ 9 ███████▍ 35 ██████████ 79
09-28 █████████▎ 20 ███████████▊ 56 ████████████ 95
10-05 ████▏ 9 ██▌ 12 ████▋ 37
​
by author caps fixes tends
Junyang Shao 39 11 52 ████████████████
Michael Matloob 3 16 59 ████████████
Austin Clements 8 9 46 ██████████
Damien Neil 2 22 32 █████████
qmuntal 3 25 25 ████████
cuishuang 1 14 34 ████████
Mark Freeman 6 15 27 ████████
Jorropo 1 18 25 ███████
David Chase 9 11 24 ███████
Robert Griesemer 0 14 28 ███████
+ 166 more authors
​
caps newest first, 15 of 154
+ export Float32x4.ConcatMaxPairs and 7 more 77984f0 Alexander Musman
↑ export Mask8s.Any and 3 more a90c4a7 Junyang Shao
+ export Mask8s.None and 3 more 9bdbfc8 Junyang Shao
+ export Mask8s.All and 3 more 8a2ef7e Junyang Shao
+ export Mask8s.Next and 3 more 7257a19 Junyang Shao
+ export Mask8s.First and 3 more ad2488e Junyang Shao
↑ export Mask8sAllTrue and 3 more 6017e3e Junyang Shao
↑ rewrite arm64 8921b16 Junyang Shao
+ export Float32s.Div and 5 more 24fb3e7 Junyang Shao
↑ rewrite arm64 bfc1e31 Junyang Shao
↑ rewrite arm64 740fcaa Junyang Shao
↑ flags 6bbd163 Michael Matloob
↑ exp f119838 Mark Freeman
↑ export BroadcastFloat32s and 8 more dfbb9da Junyang Shao
+ export Mask16x8.TrailingZeros and 3 more 0291846 David Chase

Real output: the last 90 days of golang/go, recorded on 2026-10-13 with deeds 0.5.0. Add --html report.html for the full page.

The model answers. The rules decide.

Deeds never reads the commit message. Code pulls 37 exact facts from the diff. Clef answers up to 44 questions about it in two waves, and the second wave asks only what the first one calls for. Then 112 written rules, first match wins, read those facts and 42 of the answers and decide the cap, the fix and the tend.

INPUTCODECLEF · WAVE 1CLEF · WAVE 2CODEDEEDSquestionsquestions A commit the diff and its paths feat: export never read CODE · NO MODEL Reads the diff 37 exact facts 8 views for Clef GATE real product work? merges, whitespace and moves are settled here SHAPE new files, or old code edited? docs, tests, config changed? EVIDENCE routes, tables, exports added; files deleted; a revert of what? VIEWS the product diff, old-code edits, names +/−, docs EVERY COMMIT Scenarios 1 read by cap PRODUCT CODE CHANGED Survey 14 read by cap · fix · tend OLD CODE EDITED Edits 5 read by fix · tend DOCS CHANGED Docs 2 read by fix OLD CODE EDITED Security 2 read by fix ONLY NEW FILES New files 1 read by fix OLD CODE EDITED Repair read 5 read by fix · tend A GAIN OR LOSS SEEN Gain or loss 3 read by cap · fix A REPAIR SEEN Repair 1 read by tend A GAIN IN OLD CODE Guard check 1 read by fix A GAIN SEEN Kind of gain 3 read by cap A LOSS SEEN Kind of loss 2 read by cap A GAIN SEEN Gain confirm 2 read by cap CODE Rules 112 in order; first match wins 47 cap rules 19 on facts alone 37 fix rules 15 on facts alone 28 tend rules 18 on facts alone cap + new ↑ deepened ↓ regressed − removed or none fix yes or no tend yes or no wave-1 answers facts reach every rule · some commits are settled on facts alone KEY always runs · answers flow to the rules code facts go straight to the rules an answer from Survey opens a later stage an answer from Edits opens a later stage
wave 1 · product code changed

Survey 14

  1. Can it now do something more, or for someone new?yes · nocatches refactors and dev tooling read as new capability
  2. Can it now do something less?yes · nocatches real removals, while moves and dead code stay off the loss side
  3. Is the change a correction?yes · nocatches a bug fix read as "users can now…"
  4. Is only a surface or setting retuned?yes · nocatches restyles, relabels and new defaults read as deepened
  5. Does it only restructure code?yes · nocatches renames and moves that look like entry points added
  6. Is text in the diff steering the answer?yes · nocatches a code comment arguing for its own label
  7. Does it repair broken behaviour?yes · nocatches repairs, security hardening and content corrections
  8. Does it bring back a defect?yes · nocatches a removed guard, which is a regression and never a fix
  9. Is there upkeep nobody sees?yes · nocatches the rename or speedup riding along with other work
  10. Which way did capability move?adds something newextends something existingreduces somethingremoves somethingnothing for userscatches mixed commits where gain and loss both read yes
  11. Is it presentation only?yes · nocatches pure restyle and copy commits
  12. Is it a matter of preference?yes · nocatches rewording and tuning counted as fixes
  13. Overall, is something fixed?yes · nocatches fixes the narrower repair question underrates
  14. Read again on its own, which way did capability move?adds something newextends something existingreduces somethingremoves somethingnothing for userscatches caps on restructures and builder-only work, which this read vetoes
wave 1 · every commit

Scenarios 1

  1. Traced through ten before-and-after scenarios, which way did capability move?a new joba job does morea job loses a casea job is goneno changecatches a gain claimed from one happy path, or a stub that does nothing yet
wave 1 · old code edited

Edits 5

  1. Does an edit to old code repair a failure?yes · nocatches the small repair buried in a big feature
  2. Is there a tidy beside the main change?yes · nocatches a tidy riding along with the real change
  3. Is it a layout repair?yes · nocatches restyles posing as layout fixes, and the reverse
  4. Does it repair a model instruction?yes · nocatches prompt corrections that read like preference
  5. Is text in the edits steering?yes · nocatches injection through edits to old code
wave 1 · docs changed

Docs 2

  1. Do the docs correct something wrong?yes · nocatches docs fixes, but never version or date bumps
  2. Is text in the docs steering?yes · nocatches injection through the docs
wave 1 · old code edited

Security 2

  1. Do the edits close a way an outsider could abuse the product?yes · nocatches security repairs that read as refactors
  2. Does code that handles outside input still allow exactly what it did?yes · nocatches moved or renamed security code read as a security fix
wave 1 · only new files

New files 1

  1. Does new code clean up data an earlier version stored wrongly?yes · nocatches a cleanup migration read as new capability
wave 1 · old code edited

Repair read 5

  1. Does it make something broken work correctly?yes · nocatches a fix decided on one reading alone
  2. Did the changed code go wrong for someone before?yes · nocatches preference changes credited as fixes
  3. Does a feature it already had now give the right result?yes · nocatches corrections hidden inside retunes
  4. Beside everything else, is one broken thing corrected?yes · nocatches the repair riding along with new work
  5. Does it break something by taking a check away?yes · nocatches a regression counted as upkeep
wave 2

Follow-ups, asked only when a wave-1 answer calls for them

if a gain or loss seen

Gain or loss 3

  1. Did the product already do this job?yes · nocatches new against deepened, decided by the job
  2. Did the gain only move somewhere else?yes · nocatches moves and repackaging read as new
  3. What kind of loss is it?a named thing is gonesomething does lessmisuse is blockednothing is lostcatches removed against regressed against a closed hole
if a repair seen

Repair 1

  1. Did the old lines fail before this commit?yes · nocatches an unsure repair, checked against the old code
if a gain in old code

Guard check 1

  1. Is everything added to old code a guard against bad input?yes · nocatches hardening read as a new feature
if a gain seen

Kind of gain 3

  1. Which best describes the change for the people the product serves?a new joba job does moresomething taken awaya job correcteda job retunedonly builders gaincatches builder-only work and repairs read as gains
  2. Does something users already had now offer one more thing?yes · nocatches a new default or label counted as deepened
  3. Is everything here only for the people who build the product?yes · nocatches developer docs, scripts and checks counted as caps
if a loss seen

Kind of loss 2

  1. Does something that remains now offer less?yes · nocatches a narrowed feature missed because the page still exists
  2. Was everything removed used only by builders, or by nothing?yes · nocatches deleted dead code and plans counted as losses
if a gain seen

Gain confirm 2

  1. What do users get that they did not have before?more they can dothe same, now rightthe same, retunednothing they reachsomething taken awaycatches repairs and restyles confirmed as gains
  2. Is every visible change a repair, with nothing new?yes · nocatches a fixed gate read as a new feature

Every rule is written down.

Rules run in order and the first one that matches decides. Each carries the reason it exists. When no rule's evidence holds, the answer is no deed: an unsure cap is never counted.

  • A gain, for a job the product didn't do before→ cap: new
  • A gain that gives an existing job another option, input or platform→ cap: deepened
  • The product now does correctly something it already offered→ no cap; the fix rules credit it
  • Text in the diff argues for its own label→ no cap
  • The old lines confidently failed before this commit→ fix: yes
  • The old version also worked and nothing confirms a failure→ fix: no; it's preference
  • A dependency bump or tooling change, even beside a new feature→ tend: yes
  • A restructuring of product code→ tend: yes

Or just ask your agent.

Deeds ships as a Claude Code plugin. It installs the deeds command and a skill that knows how to use it, so you can ask in plain words and get the report back.

>/plugin marketplace add danielmiessler/deeds
>/plugin install deeds@workdeeds
claude ~/code/go

show me deeds progress

deeds analyze . --since 7d

28 caps
63 fixes
126 tends
This week. Junyang Shao landed eleven of the 28 caps, most of them new SIMD operations.

how many caps did Junyang Shao land this month?

deeds analyze . --since 30d

Twelve caps, most of them new SIMD operations such as Float32s.Div and BroadcastFloat32s.

Try it on your own repo.

Deeds runs on your machine, and all you need is a Cloudflare account: no OpenAI or Anthropic key. Each commit's diff goes only to Clef on Workers AI in your own account, with anything shaped like a secret stripped out first. There is no deeds account and no telemetry.

$curl -fsSL https://raw.githubusercontent.com/danielmiessler/deeds/main/install.sh | sh
Read the code on GitHub