prdigest 0.2.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 409ae44cab5ad005c5dcb413632295fdd3aae2e8d6383852b71863f8fc484d15
-  data.tar.gz: 7ae70339af9e94123d83795362bfde4c07ee57d94bf447123fe4ce305f9117b4
+  metadata.gz: 85eefac373cdb54b93c372913fcb97cdcd4f1c003aa7cae487355ffd4da2e5e6
+  data.tar.gz: cb91e2f60e8ae566485dbdb7f56053a93b01df8fe357a9b0b6442c616c1498ec
 SHA512:
-  metadata.gz: f5c7c0e265e13c87e2a625c3ee8fd10cde644b99297277b0dd5fd1f2f4a980ea0d7d81f67a63021b9741af8811b8903af47595c9713b507dd36fb749118e4489
-  data.tar.gz: 4c787a605900e7f93722a72929708cf5032e35d1aff08a2134d41430d6e17b58f578ef20d965141391c87943eefbe3e99781a99cfcf72f08e0b35257c90d1d50
+  metadata.gz: 2c5009123d2b0df8c39612e47c89ea794a98383808cc709ddbdd1b415d32b71881733007a26f5c3dff69b13027a6c0c9e1ab59e79ee63832c363584b3ef32dc6
+  data.tar.gz: ab5fd25f659459f3dbd9a787747756bfd9f73c7cc7a56d237529eea18792a15515ac27c9f8a59e954db557338e967c3f639f7c77e318d697255e89be51307d18

data/CHANGELOG.md

@@ -1,5 +1,22 @@
 # Changelog
 
+## 0.4.0 - 2026-09-16
+
+- Add bounded PR descriptions and relevant patches to facts, plus the reusable
+  provider-free `Prdigest::Document` generation boundary for embedders.
+- Generate one concise editorial Markdown digest with dated and grouped
+  headings, user-facing outcomes, and clickable source PR links.
+- Canonicalize renamed repository aliases before collection and deduplicate
+  repeated repository configuration.
+
+## 0.3.0 - 2026-07-27
+
+- Remove `prdigest run`, deterministic Telegram HTML, schedule/cursor state,
+  catch-up processing, the `prdigest-result` envelope, and the `serve` stub.
+- Keep `prdigest facts` for agents and `prdigest prose` for stdout or
+  checkpointed plain-text Telegram delivery.
+- Make `prdigest prose --deliver` the systemd and container default.
+
 ## 0.2.0 - 2026-07-26
 
 - Add a deterministic `prdigest facts` JSON contract over the canonical

data/README.md

@@ -2,7 +2,7 @@
 
 <h1>PRDigest</h1>
 
-<p><strong>One source of truth for merged pull requests. Three ways to present it.</strong></p>
+<p><strong>One merged-PR facts contract. Prose for people and JSON for agents.</strong></p>
 
 <p>
   <a href="https://github.com/ivankuznetsov/prdigest/actions/workflows/ci.yml"><img alt="CI" src="https://github.com/ivankuznetsov/prdigest/actions/workflows/ci.yml/badge.svg"></a>
@@ -12,14 +12,14 @@
 </p>
 
 <p>
-  Collect merged PRs once, then deliver a deterministic Telegram digest, emit<br>
-  stable JSON for an agent, or generate optional prose through an
-  OpenAI-compatible Chat Completions endpoint.
+  Give OpenClaw and other agents stable merged-PR facts, or generate concise<br>
+  prose through any OpenAI-compatible Chat Completions endpoint.
 </p>
 
 <p>
   <a href="#quick-start">Quick start</a> ·
   <a href="#choose-a-mode">Choose a mode</a> ·
+  <a href="#standalone-telegram-bot">Telegram bot</a> ·
   <a href="#configuration">Configuration</a> ·
   <a href="#deployment">Deploy</a> ·
   <a href="#openclaw">OpenClaw</a>
@@ -35,20 +35,28 @@ PRDigest keeps collection separate from presentation. Every mode starts from
 the same ordered, immutable facts, so adding an agent or prose model never
 changes what was fetched from GitHub.
 
+Facts include bounded PR descriptions and relevant file patches for prose
+generation. `description_truncated`, each patch's `truncated` or `omitted`,
+and `patches_omitted` explicitly mark incomplete evidence; consumers must not
+present it as a full diff. Embedders can call
+`Prdigest::Document.generate(facts:, generator:)` with precollected facts and
+their own generator. It makes no provider, GitHub, or Telegram call itself.
+Prompt construction keeps every PR reference, title, and description, trimming
+patches first with explicit metadata and failing if non-patch metadata cannot
+fit.
+
 ```mermaid
 flowchart LR
     GH[GitHub repositories] --> C[Canonical collector]
-    C --> RUN["prdigest run<br/>deterministic Telegram"]
     C --> FACTS["prdigest facts<br/>versioned JSON"]
     C --> PROSE["prdigest prose<br/>provider-written text"]
     FACTS --> OC[OpenClaw or another client]
     PROSE --> OUT[stdout or explicit Telegram delivery]
 ```
 
-- **Deterministic by default** — repository order, pull-request order, and JSON
-  shape are stable.
-- **AI stays optional** — neither `run` nor `facts` configures or contacts a
-  prose provider.
+- **Stable facts for agents** — repository order, pull-request order, and JSON
+  shape are deterministic.
+- **AI stays explicit** — `facts` never configures or contacts a prose provider.
 - **Safe to resume** — delivery checkpoints prevent already accepted chunks
   from being sent twice.
 - **Secrets stay out of config** — YAML names environment variables; it never
@@ -58,10 +66,9 @@ flowchart LR
 
 | What you need | Command | Result | Side effects |
 |---|---|---|---|
-| Reliable scheduled digest | `prdigest run` | Deterministic Telegram HTML | Reads/writes schedule and delivery state |
-| Facts for OpenClaw or another client | `prdigest facts` | `prdigest-facts` JSON on stdout | No schedule state, Telegram, or AI provider |
+| Facts for OpenClaw or another agent | `prdigest facts` | `prdigest-facts` JSON on stdout | No Telegram, provider, or delivery state |
 | Provider-written text | `prdigest prose` | Plain text on stdout | Fresh run calls GitHub, then the provider; no Telegram or checkpoint |
-| Provider-written Telegram digest | `prdigest prose --deliver` | Checkpointed Telegram delivery | Fresh run calls GitHub and provider, persists, then sends; resume uses the checkpoint without GitHub/provider generation |
+| Scheduled or one-off Telegram prose | `prdigest prose --deliver` | Checkpointed plain-text Telegram delivery | Fresh run calls GitHub and provider, persists, then sends; resume reuses the checkpoint |
 
 All modes accept `--date YYYY-MM-DD` and repeatable
 `--repo OWNER/NAME` overrides. Repository order is always preserved.
@@ -84,7 +91,6 @@ config to state paths writable by your user:
 
 ```yaml
 state:
-  path: tmp/prdigest/state.json
   delivery_path: tmp/prdigest/deliveries
 ```
 
@@ -96,7 +102,7 @@ mode you plan to run:
 
 ```sh
 export GITHUB_TOKEN=github_pat_...
-export TELEGRAM_BOT_TOKEN=...       # run or prose --deliver only
+export TELEGRAM_BOT_TOKEN=...       # prose --deliver only
 export OPENROUTER_API_KEY=...       # prose only; use your configured env name
 ```
 
@@ -107,12 +113,9 @@ allowlisted destination chat.
 ### 2. Try the read-only paths
 
 ```sh
-# Stable JSON: no scheduler state, Telegram, or prose provider
+# Stable JSON: no Telegram, delivery state, or prose provider
 bundle exec prdigest facts --config prdigest.yml
 
-# Deterministic Telegram preview: fetches GitHub but does not send or save state
-bundle exec prdigest run --config prdigest.yml --dry-run
-
 # Provider-written text on stdout: no Telegram or delivery checkpoint
 bundle exec prdigest prose --config prdigest.yml
 ```
@@ -120,9 +123,6 @@ bundle exec prdigest prose --config prdigest.yml
 ### 3. Deliver intentionally
 
 ```sh
-# Scheduled deterministic delivery
-bundle exec prdigest run --config prdigest.yml
-
 # Provider-written delivery, checkpointed before the first Telegram request
 bundle exec prdigest prose --config prdigest.yml --deliver
 ```
@@ -131,7 +131,7 @@ To build and install the current checkout as a gem without publishing it:
 
 ```sh
 gem build prdigest.gemspec
-gem install prdigest-0.2.0.gem
+gem install prdigest-0.4.0.gem
 prdigest version
 ```
 
@@ -167,13 +167,8 @@ github:
     - owner/api
     - owner/web
 
-schedule:
-  max_catchup_days: 7
-
 digest:
   line_stats: true
-  send_empty: true
-  empty_message: "Merged PR digest — {date}\nTotal: 0 PRs"
 
 telegram:
   token_env: TELEGRAM_BOT_TOKEN
@@ -181,7 +176,6 @@ telegram:
   chat_id: -1001234567890
 
 state:
-  path: /var/lib/prdigest/state.json
   delivery_path: /var/lib/prdigest/deliveries
 
 prose:
@@ -198,12 +192,10 @@ annotated configuration.
 <summary><strong>Configuration rules</strong></summary>
 
 - Repository order controls digest order.
-- `max_catchup_days` must be between `1` and `30`.
 - `chat_id` must appear in the non-empty allowlist. Extra IDs are accepted for
   schema compatibility, but delivery sends only to `chat_id`.
 - Token values belong in environment variables, never YAML.
-- The prose block is validated only for `prdigest prose`; it does not ambiently
-  enable provider access for deterministic commands.
+- The prose block is validated only for `prdigest prose`; `facts` ignores it.
 - Remote provider URLs require HTTPS. Plaintext HTTP is accepted only for exact
   loopback hosts.
 
@@ -212,24 +204,11 @@ annotated configuration.
 ## Command reference
 
 ```text
-prdigest run [--config PATH] [--date YYYY-MM-DD] [--dry-run] [--json] [--repo OWNER/NAME ...]
 prdigest facts [--config PATH] [--date YYYY-MM-DD] [--repo OWNER/NAME ...]
 prdigest prose [--config PATH] [--date YYYY-MM-DD] [--repo OWNER/NAME ...] [--deliver]
-prdigest serve
 prdigest version
 ```
 
-### `run`
-
-An ordinary run processes owed dates oldest-first and advances state after each
-settled day. `--date` replays exactly one local date without reading or writing
-the schedule cursor. `--dry-run` previews an explicit date or yesterday and
-constructs neither state nor Telegram delivery.
-
-`--json` emits a versioned `prdigest-result` document with requested, settled,
-skipped, failed, and remaining dates plus delivery progress. `serve` is a
-compatibility stub; use the supplied systemd timer for scheduling.
-
 ### `facts`
 
 Fetches one explicit date or yesterday and prints exactly one JSON document. It
@@ -275,17 +254,17 @@ Sends the same facts document as untrusted data to
 facts without adding to or changing them.
 
 Without `--deliver`, prose is printed to stdout. With `--deliver`, the final
-escaped and chunked payload is stored under `state.delivery_path/prose` before
-the first Telegram request. Provider output containing terminal control
+plain-text chunks are stored under `state.delivery_path/prose` before the first
+Telegram request. Provider output containing terminal control
 characters is rejected before it can reach stdout, a checkpoint, or Telegram.
 
-There is no silent fallback to deterministic prose or another provider.
+There is no silent fallback to another provider.
 
 ## Delivery guarantees
 
 PRDigest treats sending as a durable protocol, not a best-effort loop:
 
-1. Fetch and render the complete day.
+1. Fetch facts and generate the complete prose digest.
 2. Persist the exact final chunk list.
 3. Mark a chunk in flight before sending.
 4. Advance only after Telegram definitely accepts it.
@@ -295,49 +274,134 @@ A definite Telegram 429/5xx response receives at most three attempts. Transport
 failures are ambiguous because Telegram may have accepted the request before
 the connection failed; PRDigest parks them instead of risking a duplicate.
 
-Prose delivery uses its own checkpoint namespace. Once a payload exists, retry
+Once a prose payload exists, retry
 loads those exact chunks before checking GitHub or provider credentials, so a
 resume never regenerates different prose. A failure before the payload becomes
 durable can incur another provider request on retry.
 
 <details>
-<summary><strong>Scheduling, replay, and state details</strong></summary>
+<summary><strong>Date and checkpoint details</strong></summary>
 
 Each local day is converted to independent UTC midnight boundaries, including
-DST gaps and repeats. A day settles only after every chunk succeeds, an enabled
-empty message succeeds, or an empty message is intentionally suppressed.
+DST gaps and repeats. Both commands use yesterday in the configured timezone
+unless `--date YYYY-MM-DD` is supplied.
 
-State is secret-free JSON version 1:
+Checkpoint directories are mode `0700`; files and locks are mode `0600`.
+A per-date lock prevents concurrent sends for the same repository scope and
+chat. While a completed checkpoint exists, repeating `prose --deliver` for that
+date is a no-op. Moving that checkpoint aside intentionally allows complete
+regeneration and redelivery.
 
-```json
-{"version":1,"timezone":"Europe/London","last_digested_date":"2026-07-15"}
+The supplied systemd timer invokes `prose --deliver` once each day. PRDigest
+does not maintain a catch-up cursor: explicitly run missed dates with `--date`.
+
+</details>
+
+## Standalone Telegram bot
+
+This path creates an outbound bot that posts one provider-written digest per
+day. It does not listen for Telegram commands; scheduling is owned by systemd,
+and every send uses `prdigest prose --deliver`.
+
+### 1. Create the bot and destination
+
+1. Open Telegram's official `@BotFather`, run `/newbot`, and keep the returned
+   token private.
+2. For a direct message, open the new bot and send `/start`. For a group, add
+   the bot and send a command such as `/start@your_bot_name` in that group.
+3. Read the destination chat ID from Telegram without putting the token in the
+   command line or browser history:
+
+   ```bash
+   read -rsp "Telegram bot token: " TELEGRAM_BOT_TOKEN
+   echo
+   export TELEGRAM_BOT_TOKEN
+   ruby -rjson -rnet/http -e '
+     uri = URI("https://api.telegram.org/bot#{ENV.fetch("TELEGRAM_BOT_TOKEN")}/getUpdates")
+     updates = JSON.parse(Net::HTTP.get(uri)).fetch("result", [])
+     keys = %w[message edited_message channel_post edited_channel_post my_chat_member chat_member]
+     chats = updates.filter_map { |update| keys.filter_map { |key| update.dig(key, "chat") }.first }
+     chats.uniq { |chat| chat.fetch("id") }.each do |chat|
+       puts [chat.fetch("id"), chat["title"] || chat["username"] || chat["first_name"]].compact.join("\t")
+     end
+   '
+   unset TELEGRAM_BOT_TOKEN
+   ```
+
+If no ID appears, send the bot another command and repeat the lookup. Direct
+chat IDs are normally positive; groups and channels normally use negative IDs.
+
+### 2. Install and configure PRDigest
+
+Install the released gem:
+
+```sh
+gem install prdigest -v 0.4.0
+prdigest version
+```
+
+Create a private `prdigest.yml` with the repositories, destination chat, and
+provider you want. This example uses OpenRouter with DeepSeek V4 Flash:
+
+```yaml
+timezone: Europe/London
+
+github:
+  token_env: GITHUB_TOKEN
+  repos:
+    - owner/api
+    - owner/web
+
+digest:
+  line_stats: true
+
+telegram:
+  token_env: TELEGRAM_BOT_TOKEN
+  chat_id_allowlist: [-1001234567890]
+  chat_id: -1001234567890
+
+prose:
+  provider: openai_compatible
+  base_url: https://openrouter.ai/api/v1
+  model: deepseek/deepseek-v4-flash
+  api_key_env: OPENROUTER_API_KEY
 ```
 
-Writes use atomic mode-`0600` replacement and directory fsync. Delivery
-directories are mode `0700`; files and locks are mode `0600`. A per-date lock
-prevents concurrent sends for the same checkpoint.
+The provider and model above are only an example. Replace them with any
+OpenAI-compatible endpoint and model you prefer.
 
-Missing state means first run and requests yesterday only. Malformed, future,
-unsupported, unreadable, or timezone-mismatched state fails closed.
+Replace both chat IDs with the value from step 1. Keep the GitHub, Telegram,
+and provider tokens in environment variables, never in YAML. Use a dedicated
+bot token, a read-only fine-grained GitHub token, and a provider key with a
+small spend limit.
 
-When backlog exceeds the configured cap, PRDigest durably skips the oldest
-prefix and processes only the newest window. For a timezone migration, stop the
-timer, preserve and move aside the old state, change timezone, explicitly
-replay any required dates, then restart the timer.
+### 3. Preview, send once, then schedule
 
-GitHub does not guarantee search-index freshness. Keep the host timezone aligned
-with the digest timezone and use `--date YYYY-MM-DD` for a delayed merge found
-by a later audit. Explicit replay uses the same delivery ledger: while a
-completed checkpoint for that date exists, delivery is a no-op and newly
-rendered content is ignored. Intentionally archiving the checkpoint makes the
-next replay regenerate and send the complete date, not only the delayed pull
-request.
+Read the three credentials without saving them in shell history, preview the
+prose without Telegram, then perform the first intentional delivery:
 
-</details>
+```bash
+read -rsp "GitHub token: " GITHUB_TOKEN; echo
+read -rsp "Telegram bot token: " TELEGRAM_BOT_TOKEN; echo
+read -rsp "OpenRouter API key: " OPENROUTER_API_KEY; echo
+export GITHUB_TOKEN TELEGRAM_BOT_TOKEN OPENROUTER_API_KEY
+
+prdigest prose --config ./prdigest.yml
+prdigest prose --config ./prdigest.yml --deliver
+unset GITHUB_TOKEN TELEGRAM_BOT_TOKEN OPENROUTER_API_KEY
+```
+
+Once the one-off send succeeds, use the tested [systemd deployment](#deployment)
+below. Its timer runs at `09:05` in the host timezone and sends the previous
+day's digest. Change `OnCalendar` before enabling the timer if you want another
+time. Re-running a completed date is a no-op; use `--date YYYY-MM-DD` for an
+explicit missed day.
 
 ## OpenClaw
 
-The repository includes a ClawHub-ready skill at
+The repository includes the published ClawHub skill
+[`@ivankuznetsov/prdigest`](https://clawhub.ai/ivankuznetsov/skills/prdigest)
+under **Development**, with its source at
 [`openclaw/skills/prdigest/SKILL.md`](openclaw/skills/prdigest/SKILL.md). It:
 
 - invokes only `prdigest facts`;
@@ -346,31 +410,53 @@ The repository includes a ClawHub-ready skill at
 - makes no second GitHub request; and
 - never invokes delivery or configures a prose provider.
 
-The skill is source-ready, **not claimed as published**. After a separately
-authorized ClawHub release, the expected command is:
+The Ruby CLI and ClawHub skill are separate installs with independent version
+histories. Pin the Ruby gem release you want, but let ClawHub resolve the
+current skill. To let OpenClaw install both, copy and paste this prompt into an
+OpenClaw chat:
 
-```sh
-clawhub install @ivankuznetsov/prdigest
+```text
+Install PRDigest 0.4.0 in the same user/runtime context as OpenClaw with
+`gem install prdigest -v 0.4.0`, then install the ClawHub skill with
+`openclaw skills install @ivankuznetsov/prdigest`. This message
+explicitly authorizes those two installs and only the PATH adjustment needed to
+make the installed `prdigest` executable visible to the OpenClaw runtime. Do
+not create PRDigest configuration files, store credentials, enable Telegram
+delivery, or install a scheduler. First verify Ruby 3.2 or newer is available;
+if it is not, stop and report the exact blocker instead of changing system
+packages. After installation, run `prdigest version`, confirm that OpenClaw can
+discover the installed PRDigest skill, and report the installed paths and
+versions without exposing environment variables or tokens.
 ```
 
-ClawHub installation and Ruby CLI installation are separate trust boundaries.
-See [`openclaw/README.md`](openclaw/README.md) for local use and the
-release-gated publication checklist.
+For a manual install, run `gem install prdigest -v 0.4.0` and
+`openclaw skills install @ivankuznetsov/prdigest`. The skill gives an agent
+facts-to-prose behavior only; use the
+[standalone Telegram bot](#standalone-telegram-bot) when PRDigest itself should
+generate and deliver scheduled prose.
+
+See [`openclaw/README.md`](openclaw/README.md) for local development and
+publication details.
 
 ## Deployment
 
 <details open>
-<summary><strong>systemd on Ubuntu</strong></summary>
+<summary><strong>systemd on Linux</strong></summary>
 
-Install Ruby, `tzdata`, and the gem, then:
+Install Ruby 3.2 or newer and `tzdata`, then install the gem system-wide. The
+gem contains the example configuration and tested service units, so a source
+checkout is not required:
 
 ```sh
-sudo useradd --system --home /nonexistent --shell /usr/sbin/nologin prdigest
+sudo gem install prdigest -v 0.4.0 --no-document --bindir /usr/local/bin
+gem_root=$(ruby -e 'print Gem::Specification.find_by_name("prdigest", "0.4.0").full_gem_path')
+
+sudo useradd --system --home /nonexistent --shell "$(command -v nologin)" prdigest
 sudo install -d -o root -g prdigest -m 0750 /etc/prdigest
-sudo install -o root -g prdigest -m 0640 configs/config.example.yml /etc/prdigest/config.yml
-sudo install -o root -g root -m 0600 .env.example /etc/prdigest/.env
-sudo install -o root -g root -m 0644 scripts/systemd/prdigest.service /etc/systemd/system/
-sudo install -o root -g root -m 0644 scripts/systemd/prdigest.timer /etc/systemd/system/
+sudo install -o root -g prdigest -m 0640 "$gem_root/configs/config.example.yml" /etc/prdigest/config.yml
+sudo install -o root -g root -m 0600 /dev/null /etc/prdigest/.env
+sudo install -o root -g root -m 0644 "$gem_root/scripts/systemd/prdigest.service" /etc/systemd/system/
+sudo install -o root -g root -m 0644 "$gem_root/scripts/systemd/prdigest.timer" /etc/systemd/system/
 sudoedit /etc/prdigest/config.yml
 sudoedit /etc/prdigest/.env
 sudo systemctl daemon-reload
@@ -378,6 +464,10 @@ sudo systemctl start prdigest.service
 sudo systemctl enable --now prdigest.timer
 ```
 
+Put `GITHUB_TOKEN=...`, `TELEGRAM_BOT_TOKEN=...`, and the configured provider
+key such as `OPENROUTER_API_KEY=...` in `/etc/prdigest/.env`. Do not prefix
+systemd environment-file entries with `export`.
+
 systemd creates `/var/lib/prdigest` as `prdigest:prdigest` mode `0700`.
 
 ```sh
@@ -416,25 +506,24 @@ docker run --rm --env-file /etc/prdigest/.env \
 ### Rollback
 
 Stop the timer, install the prior gem or image, restore its matching config and
-a known-good state backup, then restart. Replay omitted dates explicitly; never
-move a checkpoint forward by hand.
+checkpoint backup, then restart. Never move a checkpoint forward by hand.
 
 ## Exit codes and troubleshooting
 
 | Exit | Meaning | First check |
 |---:|---|---|
-| `0` | Completed or dry-run | — |
+| `0` | Completed | — |
 | `1` | Unexpected or render failure | Logs and input shape |
 | `2` | CLI/configuration refusal | Config path, YAML, timezone, allowlist, env |
 | `3` | GitHub failure | Token scope, repository access, rate/search limits |
 | `4` | Telegram failure | `error.kind` and delivery checkpoint |
-| `5` | State failure | Path, owner, mode, JSON version/date/timezone |
-| `6` | Failure after durable progress | Settled/skipped dates before retry |
+| `5` | Checkpoint state failure | Path, owner, mode, checkpoint JSON |
 | `7` | Provider failure or ambiguous outcome | Endpoint, model, key env, retry cost |
 
 Reconcile `telegram_ambiguous`, `telegram_permanent`, and
-`delivery_checkpoint_permanent` before moving a checkpoint. Concurrent
-scheduled runs are unsupported; the systemd oneshot is the normal coordinator.
+`delivery_checkpoint_permanent` before moving a checkpoint. Overlapping delivery
+for the same date is refused by the checkpoint lock; the systemd oneshot is the
+normal scheduler.
 
 See [`SECURITY.md`](SECURITY.md) for token scope, rotation, and private-data
 flow.

data/SECURITY.md

@@ -35,8 +35,8 @@ and model boundary so OpenClaw can write prose. Standalone `prdigest prose`
 sends the same complete document plus the configured model name to the
 OpenAI-compatible endpoint. Private repositories therefore require an OpenClaw
 deployment or provider whose data handling, retention, and access controls are
-acceptable to the operator. Neither mode is ambient: `prdigest run` and
-`prdigest facts` never contact the standalone prose provider.
+acceptable to the operator. Provider access is never ambient: `prdigest facts`
+does not contact the standalone prose provider.
 
 Pull-request fields are untrusted input. The built-in provider prompt and the
 OpenClaw skill explicitly classify the facts JSON as data, never instructions,
@@ -45,9 +45,8 @@ processing private repository content. The OpenClaw skill must not make a second
 GitHub query, deliver messages, install software silently, or print credentials.
 
 Prose Telegram checkpoints contain generated text derived from repository facts.
-They use the same secret-free mode-`0600` files and mode-`0700` directories as
-deterministic delivery, under a separate `prose` namespace. Protect, retain, and
-delete them as private repository metadata.
+They use secret-free mode-`0600` files inside mode-`0700` directories. Protect,
+retain, and delete them as private repository metadata.
 
 ## Reporting
 

data/configs/config.example.yml

@@ -1,12 +1,6 @@
-# prdigest — multi-repo merged-PR daily digest (Hive-independent)
+# prdigest — merged-PR facts for agents and provider-written prose
 
 timezone: Europe/London
-schedule:
-  # Prefer systemd timer + `prdigest run` for v1.
-  # serve mode may honor this later.
-  cron: "5 9 * * *"
-  # Newest retained catch-up window; valid range is 1..30.
-  max_catchup_days: 7
 
 github:
   token_env: GITHUB_TOKEN
@@ -23,11 +17,9 @@ telegram:
 
 digest:
   line_stats: true
-  send_empty: true
-  empty_message: "Merged PR digest — {date}\nTotal: 0 PRs"
 
 prose:
-  # Used only by `prdigest prose`; `run` and `facts` never contact this endpoint.
+  # Used only by `prdigest prose`; `facts` never contacts this endpoint.
   provider: openai_compatible
   base_url: https://openrouter.ai/api/v1
   # Replace with a model identifier accepted by the configured endpoint.
@@ -36,6 +28,5 @@ prose:
   api_key_env: OPENROUTER_API_KEY
 
 state:
-  path: /var/lib/prdigest/state.json
-  # Stable rendered chunks and next-unsent position, separate from the date cursor.
+  # Stable prose chunks and the next-unsent position.
   delivery_path: /var/lib/prdigest/deliveries