Agent skill

Call Temporal Anchor Auditor

by CALLE-AI in CALLE-AI/awesome-phone-call-agents

Offline experimental CALL-E transcript helper that resolves every time reference in the agent's turns against the call timestamp, flagging relative-only commitments, meridiem-less clocks…

MITAuto-check passed

Install Call Temporal Anchor Auditor

skills CLI
$ npx skills add CALLE-AI/awesome-phone-call-agents --skill call-temporal-anchor-auditor -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install CALLE-AI/awesome-phone-call-agents call-temporal-anchor-auditor --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/CALLE-AI/awesome-phone-call-agents.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/call-temporal-anchor-auditor .claude/skills/call-temporal-anchor-auditor && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
call-temporal-anchor-auditor
GitHub stars
107
Token cost
~2.6k tokens
SKILL.md length
1,334 words
Files
9 (incl. scripts, references)
Skills in repo
101
Repo updated
First seen
Licence
MIT

At a glance

Offline experimental CALL-E transcript helper that resolves every time reference in the agent's turns against the call timestamp, flagging relative-only commitments, meridiem-less clocks…

  • SKILL.md covers When To Use, When Not To Use, Verdicts and How It Works, plus 4 more sections
  • Runs Python scripts from its folder; calls python3

What it does

Call Temporal Anchor Auditor is an agent skill from CALLE-AI/awesome-phone-call-agents. Offline experimental CALL-E transcript helper that resolves every time reference in the agent's turns against the call timestamp, flagging relative-only commitments, meridiem-less clocks, dialect-dependent weekday phrases, and weekday-date calendar conflicts, plus an absolute-time goal template. It is not proof the callee misunderstood, a scheduler, or authorization to act.

Its SKILL.md is about 2.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 10 other files, including scripts and reference files (for example `references/example-transcript-conflict.json`, `references/example-transcript-relative.json` and `references/example-transcript.json`).

The repository describes itself as: Portable phone-call Agent Skills, apps, examples, adapters, and scheduler recipes for AI agents. The licence is MIT.

Example prompts

  • “/call-temporal-anchor-auditor”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 38d4118. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 2 files in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • iso.org

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Call Temporal Anchor Auditor loads about 2.6k tokens when it runs, and up to ~5.6k if it reads all its reference files. Until then it costs about 101 tokens; SKILL.md has 1,334 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~101
When it runs · the whole SKILL.md, loaded when a task matches
~2.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~5.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from CALLE-AI/awesome-phone-call-agents at commit 38d4118, republished under its MIT licence (© CALLE-AI). 1,334 words, ~2,610 tokens.

Download SKILL.mdSave it as .claude/skills/call-temporal-anchor-auditor/SKILL.md (or your agent's skills folder). This skill also uses 8 other files; get the full folder from GitHub.
name
call-temporal-anchor-auditor
description
Offline experimental CALL-E transcript helper that resolves every time reference in the agent's turns against the call timestamp, flagging relative-only commitments, meridiem-less clocks, dialect-dependent weekday phrases, and weekday-date calendar conflicts, plus an absolute-time goal template. It is not proof the callee misunderstood, a scheduler, or authorization to act.
license
MIT

call-temporal-anchor-auditor

A promise with no calendar anchor is a no-show waiting to happen.

CALL-E's highest-value calls are bookings: appointment-confirm, no-show-shield, slotsaver, capacity-backfill-cascade all live or die on whether the person on the other end walks in at the right slot. An agent that closes with "see you tomorrow evening" or "your pickup is Tuesday at 2" has exchanged a calendar anchor for a vibe - and each unanchored minute is paid for later as a missed appointment, an idle bay, or a double-booked table. This skill grades every time reference in the agent's own turns and tells you which commitments would not survive a calendar check.

When To Use

  • after any booking, confirmation, or appointment call, ideally with --called-at set to the run's ISO-8601 start time so relative phrases ("tomorrow", "next month") resolve against a real reference date
  • whenever the transcript shows commitments but you cannot say, from the text alone, which calendar day and clock time the agent actually locked in
  • before placing the next wave of calls, to craft a plan_call goal that forces absolute-time discipline (weekday + date + clock + meridiem)

When Not To Use

  • without --called-at when the relative phrases matter: without a reference time they grade unresolvable_without_call_time and weekday-date pair checks are skipped - if you know when the call ran, pass the flag
  • on non-English transcripts; the temporal lexicons are English-only
  • to prove the callee misunderstood the time; the card grades the agent's wording only, never the callee's comprehension or intent
  • during a call; strictly post-call analysis plus pre-call goal crafting

Verdicts

VerdictMeaning
NO_TIME_REFERENCESthe agent's turns contain no detectable time expression at all
INTERNAL_DATE_CONFLICTthe agent's own turns disagree with the calendar (stated weekday vs date) or restate a clock hour with a different value
RELATIVE_ONLY_COMMITMENTSa commitment turn carries time expressions but never reaches a full absolute anchor anywhere the call supports it
AMBIGUOUS_TIME_REFERENCEStime references exist and no commitment failed, but ambiguous ones remain (dialect-dependent weekdays, day-part bands, meridiem-less clocks, week ranges)
FULLY_ANCHOREDevery committed time resolves to an absolute date and a meridiem-bearing clock time

How It Works

bash
python3 scripts/call_temporal_anchor_auditor.py analyze \
  --call-result path/to/call-result.json \
  --called-at 2026-10-07T09:00:00-07:00

Reads the real get_call_run result shape ({status, result: {transcript}}) or a bare JSON array of {speaker, text} turns. Only AGENT turns are audited; callee time mentions are counted but never graded.

Collection classes:

  • absolute_date: "October 14", "the 14th of October", "10/14" (US MM/DD), and bare ordinal days ("the 14th") whose month is inferred from the call timestamp
  • clock_absolute: clock times with meridiem ("at 2 p.m."), canonicalized to 24h values; meridiem-less forms are clock_ambiguous - "at 2" and colon times at or below 12 ("at 2:30") could be morning or afternoon, while 13-23 colon times ("14:30") are true 24-hour clocks
  • invalid_date: a calendar date that does not exist in the call year ("February 29" in 2026) - never satisfies anchoring
  • band: day-part phrases ("in the evening", "tomorrow morning", "at noon")
    • vague on their own, no clock inside
  • relative_resolved / relative_derived: "tomorrow", "in two weeks", "a week from today", bare weekdays - resolved only with --called-at; a bare weekday resolves to the next strictly future occurrence and is never guessed for "next/this <weekday>", which is flagged ambiguous (dialect-dependent)
  • unresolvable_without_call_time: relatives and bare ordinal days seen with no --called-at

Resolution rules worth knowing:

  • "next month" resolves to the same day next month, clamped to the target month's end (Jan 31 -> Feb 28 in non-leap years)
  • a bare weekday never resolves to today; strictly future, at least tomorrow
  • "next <weekday>" and "this <weekday>" are never resolved; their meaning varies by dialect, so they always flag

Conflict rules (any one forces INTERNAL_DATE_CONFLICT):

  • (a) a weekday+date pair in one turn contradicts the calendar of the --called-at year ("Tuesday, October 14" when Oct 14 is a Wednesday)
  • (b) the same resolved date is stated with different weekdays across turns
  • (c) the same 12-hour clock number is restated with a different value or meridiem ("at 2" plus "at 3 p.m." both keyed to hour 12)

Commitment anchoring: turns matching commitment verbs (confirm, reserve, book, see you, pickup, appointment, ...) that carry time expressions get a RELATIVE_ONLY_COMMITMENT finding unless the slot reaches a full anchor - a date anchor plus a clock_absolute - which may live anywhere in the call; anchoring is call-level, so a restatement can lean on the anchor stated earlier.

Craft the absolute-time goal
bash
python3 scripts/call_temporal_anchor_auditor.py craft \
  --task "confirm the catering pickup" \
  --date "Wednesday, October 14" --time "2 p.m."

Emits a plan_call goal whose time-discipline block demands weekday + calendar date + clock time + meridiem on every commitment, forbids bare "next <weekday>" and dateless "tomorrow", and requires the full anchor to be read back once. See references/example-goal.txt for a realistic result.

Show full SKILL.md (588 more words)Show less

Limitations

  • Evidence text fields for clock times can render with the trailing dot trimmed ("at 2 p.m") while the value stays canonical ("14:00"); judge anchoring by values, not by the echoed text.
  • Clock conflict rule (c) keys on the 12-hour number, so "open at 9 a.m." plus "appointment at 9 p.m." in one call flags an INTERNAL_CLOCK_CONFLICT even though both statements may be correct. Treat clock conflicts as a prompt to read the turns, not as proof of a contradiction.
  • The PII masker hides numeric dates with 7+ total digits before analysis ("10/14/2026" becomes "#" padding), so fully numeric ISO-style dates are invisible to the auditor. State dates as month names ("October 14") in goals and verify such dates manually.
  • Slash dates assume US MM/DD ordering; "10/14" is October 14, never 10 October in a DD/MM reading.
  • "this/next week" and "this/next weekend" classify as ambiguous with reason "week-range reference without a concrete day" even when --called-at is given; a week range never resolves to a day.
  • A weekday preceded by an ordinal or frequency word ("first Friday", "every Tuesday", "last Monday") classifies as ambiguous with reason "complex ordinal weekday expression"; the "nth weekday of a month" computation is dialect- and calendar-dependent, so it is flagged, never resolved.
  • Commitment anchoring is call-level: an anchor anywhere in the call satisfies every commitment turn, so a locally vague restatement of an already-anchored slot will not flag.

Testing

bash
python3 -m pytest skills/call-temporal-anchor-auditor -q

81 tests cover the collector, resolver, conflict rules, commitment findings, verdict priority, CLI battery, craft mode, and the shipped fixtures. references/examples.md shows byte-real runs against them.

Research & Standards Grounding

ResearchRelevance
Fatemi, Kazemi, Tsitsulin, Malkan, Yim, Palowitch, Seo, Halcrow, Perozzi. "Test of Time: A Benchmark for Evaluating LLMs on Temporal Reasoning". arXiv:2406.09170 (2024).Large-scale evidence that temporal reasoning is a measured LLM weakness across order, duration, and absolute/relative references - the failure class this skill audits on call transcripts
Su, Li, Zhang, Zhu, Qu, Zhou, Bowen, Cheng, Zhang. "Living in the Moment: Can Large Language Models Grasp Co-Temporal Reasoning?" (CoTempQA). ACL 2024. arXiv:2406.09072.Shows co-temporal reasoning (aligning statements with a reference time) degrades even in strong models - exactly the resolution this skill performs deterministically against --called-at
Wei, Li, Song, Luo, Zhuang, Tan, Guo, Wang. "TIME: A Multi-level Benchmark for Temporal Reasoning of LLMs in Real-World Scenarios" (includes TIME-Dial, a dialogue sub-benchmark). NeurIPS 2025 (Spotlight). arXiv:2505.12891.Demonstrates the weakness persists in dialogue settings (TIME-Dial), the register CALL-E transcripts live in
ISO 24617-1:2012 "Language resource management - Semantic annotation framework (SemAF) - Part 1: Time and events (SemAF-Time, ISO-TimeML)". International standard, https://www.iso.org/standard/37331.html (confirmed current by its 2023 systematic review).The standards model for annotating absolute vs relative time expressions and resolving them against a reference time - this skill's collection classes follow the same shape as deterministic heuristics

Together the three 2024-2025 benchmarks evidence that temporal reasoning - including in dialogue (TIME-Dial) - is a measured LLM weakness, and ISO-TimeML is the standards model for absolute vs relative time expressions resolved against a reference time; this skill operationalizes both as deterministic transcript heuristics, not as a model judgment.

Citation note: the Test of Time author list is nine names with Fatemi as first author; some secondary sources misattribute the paper, so cite Fatemi.

Differences from sibling skills

  • call-cross-call-consistency-checker compares stated facts across two calls; this skill compares the agent's time wording against the calendar within one call.
  • call-agent-commitment-tracker inventories what was promised; this skill grades whether the promised times would survive a calendar check.
  • appointment-confirm and no-show-shield act on bookings; this skill is a read-only post-call audit that feeds them a verifiable anchor.

© CALLE-AI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 8 other files (scripts, references) in skills/call-temporal-anchor-auditor of CALLE-AI/awesome-phone-call-agents.

  • SKILL.md
  • references/example-goal.txt
  • references/example-transcript-conflict.json
  • references/example-transcript-relative.json
  • references/example-transcript.json
  • references/examples.md
  • references/safety.md
  • scripts/call_temporal_anchor_auditor.py
  • scripts/test_call_temporal_anchor_auditor.py

Open the folder on GitHubat commit 38d4118

Compare with similar skills

Call Temporal Anchor Auditor next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Call Temporal Anchor Auditor compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Call Temporal Anchor Auditor this skillCALLE-AI/awesome-phone-call-agents107—~2.6kAutomated safety check: PassMIT
Baoyu Youtube TranscriptJimLiu/baoyu-skills27k1 repos~2.4kAutomated safety check: PassMIT
Youtube Transcriptbrowser-act/skills6.1k—~2.1kAutomated safety check: PassMIT
Temporal Date Window ResolverHarnessMD/munder-difflin8.6k—~742Automated safety check: NotesMIT
Youtube Transcriptsickn33/agentic-awesome-skills47k1 repos~1.3kAutomated safety check: PassMIT
Transcription0xsline/OpenChatCut2.2k1 repos~1.1kAutomated safety check: PassAGPL-3.0

Similar skills

  • Baoyu Youtube Transcript

    JimLiu/baoyu-skills

    Downloads YouTube video transcripts/subtitles and cover images by URL or video ID.

    27k GitHub starsUsed in 1 repo~2.4k tokens
    Media & CreativeAuto-check passed
  • Youtube Transcript

    browser-act/skills

    YouTube transcript extraction and content reformatting: given a YouTube video URL, opens the video's transcript panel, extracts all timestamped segments, and transforms the raw transcript into…

    6.1k GitHub stars~2.1k tokensUpdated 1 mo ago
    Knowledge ManagementAuto-check passed
  • Temporal Date Window Resolver

    HarnessMD/munder-difflin

    Resolves a named time window like lastWeek, last30Days or lastQuarter into exact ISO start and end dates relative to the current moment, with no files written and no network calls.

    8.6k GitHub stars~742 tokensUpdated yesterday
    DevelopmentAuto-check: notes
  • Youtube Transcript

    sickn33/agentic-awesome-skills

    Fetch YouTube transcripts through DeepAPI or local fallback tooling and save clean text output.

    47k GitHub starsUsed in 1 repo~1.3k tokens
    Knowledge ManagementAuto-check passed
  • Transcription

    0xsline/OpenChatCut

    A skill your agent uses when a video/audio task needs OpenChatCut transcription, captions, subtitles, subtitle styling, transcript search, transcript readiness checks, or enabling captions…

    2.2k GitHub starsUsed in 1 repo~1.1k tokens
    Media & CreativeAuto-check passed
  • Youtube Transcript Skills

    sickn33/agentic-awesome-skills

    Fetch YouTube video transcripts, search videos/channels, browse channels, and extract playlists via the getyoutubetranscript.com API - free tier, no card required.

    47k GitHub starsUsed in 1 repo~1.2k tokens
    Knowledge ManagementAuto-check passed

More from CALLE-AI/awesome-phone-call-agents

All 101 skills in this repo
  • Accessible Outing Verifier

    CALLE-AI/awesome-phone-call-agents

    Demonstrates advisory accessibility-planning checks with offline fixtures and a proposed bounded CALL-E workflow; use for exploring unknown or qualified venue claims without making calls.

    107 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed
  • Ground Truth Gate

    CALLE-AI/awesome-phone-call-agents

    A skill your agent uses when an agent holds some evidence for a physical-world claim but the evidence is broader, narrower, or older than the exact question asked, and it must first decide whether a…

    107 GitHub stars~3.3k tokensUpdated yesterday
    Auto-check passed
  • Is It Accessible

    CALLE-AI/awesome-phone-call-agents

    Call a venue and ask the accessibility questions that matter to one specific person — step-free entry, hearing loop, guide dogs, quiet hours, changing places — then return a per-need verdict backed…

    107 GitHub stars~4.3k tokensUpdated yesterday
    Auto-check passed
  • Landmark Navigation Assist

    CALLE-AI/awesome-phone-call-agents

    Turns a pre-written, building-level location config into a CALL-E outbound phone-call task that guides a delivery driver through the last few hundred metres to a specific building using landmarks…

    107 GitHub stars~1.6k tokensUpdated yesterday
    Auto-check passed
  • Research Gap Call Verifier

    CALLE-AI/awesome-phone-call-agents

    Turn cited business research into a bounded, approval-gated phone-call plan that asks only unresolved factual questions, then reconcile CALL-E-compatible results without treating voicemail, refusal…

    107 GitHub stars~1.7k tokensUpdated yesterday
    Auto-check passed
  • Structured Outcome Followup Call

    CALLE-AI/awesome-phone-call-agents

    Place a goal-driven CALL-E call that collects specific structured answers, score those answers against a deterministic rubric you supply, and conditionally trigger a follow-up action — all runnable…

    107 GitHub stars~1.4k tokensUpdated yesterday
    Auto-check passed

Questions about Call Temporal Anchor Auditor

What does Call Temporal Anchor Auditor do?

Offline experimental CALL-E transcript helper that resolves every time reference in the agent's turns against the call timestamp, flagging relative-only commitments, meridiem-less clocks…. Call Temporal Anchor Auditor is an agent skill from CALLE-AI/awesome-phone-call-agents. Offline experimental CALL-E transcript helper that resolves every time reference in the agent's turns against the call timestamp, flagging relative-only commitments, meridiem-less clocks, dialect-dependent weekday phrases, and weekday-date calendar conflicts, plus an absolute-time goal template.

How do I install Call Temporal Anchor Auditor in Claude Code?

Run `npx skills add CALLE-AI/awesome-phone-call-agents --skill call-temporal-anchor-auditor -a claude-code`. Or copy the skill folder (skills/call-temporal-anchor-auditor in CALLE-AI/awesome-phone-call-agents) into .claude/skills/call-temporal-anchor-auditor in your project. Claude Code loads it when a task matches its description.

How do I install Call Temporal Anchor Auditor in Codex?

Run `npx skills add CALLE-AI/awesome-phone-call-agents --skill call-temporal-anchor-auditor -a codex`. Or copy the skill folder (skills/call-temporal-anchor-auditor in CALLE-AI/awesome-phone-call-agents) into .agents/skills/call-temporal-anchor-auditor in your project. Codex loads it when a task matches its description.

Can I use Call Temporal Anchor Auditor in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add CALLE-AI/awesome-phone-call-agents --skill call-temporal-anchor-auditor -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/call-temporal-anchor-auditor, .gemini/skills/call-temporal-anchor-auditor, .github/skills/call-temporal-anchor-auditor and .opencode/skills/call-temporal-anchor-auditor in your project.

What does Call Temporal Anchor Auditor need to run?

Going by SKILL.md and its folder, Call Temporal Anchor Auditor needs Python for the scripts in its folder and the command-line tools its instructions call (python3). Our summary lists: Python 3.

Does Call Temporal Anchor Auditor access the network?

SKILL.md names 1 domain. As links in the text: iso.org. This is read from the text; nothing was executed.

Is Call Temporal Anchor Auditor safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Call Temporal Anchor Auditor use?

Call Temporal Anchor Auditor is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Call Temporal Anchor Auditor use?

About 2.6k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 3k tokens, read only when the agent opens those files.

What are the alternatives to Call Temporal Anchor Auditor?

Skills that share tags, products or a category with Call Temporal Anchor Auditor: Baoyu Youtube Transcript (JimLiu/baoyu-skills, 27k stars), Youtube Transcript (browser-act/skills, 6.1k stars), Temporal Date Window Resolver (HarnessMD/munder-difflin, 8.6k stars) and Youtube Transcript (sickn33/agentic-awesome-skills, 47k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Call Temporal Anchor Auditor?

CALLE-AI (a GitHub organization) maintains it in CALLE-AI/awesome-phone-call-agents, which has 107 GitHub stars. The repository holds 101 skills in this directory. The repository was last updated on October 10, 2026.

Source: CALLE-AI/awesome-phone-call-agents on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.